Biyani's Think Tank

Concept based notes

Advanced Computer Architecture

(BCA-III Year)

Nitika Newar, MCA

Revised By: Namarta
Dept. of Information Technology
Biyani Girls College, Jaipur

Published by :

Think Tanks
Biyani Group of Colleges

Concept & Copyright :

Biyani Shikshan Samiti

Sector-3, Vidhyadhar Nagar,
Jaipur-302 023 (Rajasthan)
Ph : 0141-2338371, 2338591-95 Fax : 0141-2338007
E-mail : [email protected]
Website :www.gurukpo.com; www.biyanicolleges.org
ISBN: 978-93-82801-79-5

Edition : 2011
Price :
While every effort is taken to avoid errors or omissions in this Publication, any mistake or
omission that may have crept in is not intentional. It may be taken note of that neither the
publisher nor the author will be responsible for any damage or loss of any kind arising to
anyone in any manner on account of such errors and omissions.

Leaser Type Setted by :

Biyani College Printing Department

Advanced Computer Arc.

Preface

am glad to present this book, especially designed to serve the needs of the

students. The book has been written keeping in mind the general weakness in
understanding the fundamental concepts of the topics. The book is self-explanatory and
adopts the Teach Yourself style. It is based on question-answer pattern. The language of
book is quite easy and understandable based on scientific approach.
Any further improvement in the contents of the book by making corrections, omission
and inclusion is keen to be achieved based on suggestions from the readers for which the
author shall be obliged.
I acknowledge special thanks to Mr. Rajeev Biyani, Chairman & Dr. Sanjay Biyani,
Director (Acad.) Biyani Group of Colleges, who are the backbones and main concept
provider and also have been constant source of motivation throughout this Endeavour. They
played an active role in coordinating the various stages of this Endeavour and spearheaded
the publishing work.
I look forward to receiving valuable suggestions from professors of various educational
institutions, other faculty members and students for improvement of the quality of the book.
The reader may feel free to send in their comments and suggestions to the under mentioned
address.
Author

Syllabus
B.C.A. Part-III

Advanced Computer Architecture

Parallel Computer Models : The state of computing, multiprocessors and
multicomputer, multivector and SIMD computers, architectural development
tracks.
Program and Network Properties : Conditions of parallelism, program
partitioning and scheduling, program flow mechanisms.
System Interconnect Architectures : Network properties and routing, Static
interconnection network and dynamic intercommection networks.
Processors and Memory Hierachy : Advanced processor technologyCISC,
RISC, Superscalar, Vector VLIW and symbolic processors, memory
technology.
Bus, Cache and Shared Memory.
Linear Pipeline Processors, Nonlinear Pipeline, processors Instruction
pipeline Design Multiprocessors System Interconnects Vector Processing
Principles, Multivector Multiprocessors.

Advanced Computer Arc.

Content

S. No.

Name of Topic

1.

Parallel Computer Models

1.1
Multiprocesors
1.2
Parallel processing
1.3
State of computing
1.4
History of computer Architecture
1.5
Parallelism
1.6
Levels of Paralleism
1.7
Vector super computers
1.8
Shared memory multiprocessor
1.9
Distributed memory multicomputers
1.10 SIMD computers
1.11 Architectural Development Tracks
1.12 SIMD array processor

2.

Program partitioning or scheduling

2.1
Program Flow Mechanisms
2.2
Data flow Architecture
2.3
Grain Sizes & Latency
2.4
Scheduling procedure

3.

System Interconnect Architecture

3.1
Network properties
3.2
Bisection width
3.3
Data routing functions
3.4
Static interconnection networks
3.5
Dynamic interconnection networks

4.

Processors and Memory Hierarchy

4.1
RISC & CISC
4.2
Super scale processors
4.3
VLIW Architecture

5.
5.1

6.

4.4
Symbolic Processors
4.5
Virtual Memory
4.6
Backplane Bus
Pipeline processors
Characteristics of Pipeline
5.2
Linear pipeline processors
5.3
Non Linear pipeline processors
5.4
Reservation table in linear pipelining & non
linear pipelining
5.5.
Instruction pipeline design
5.6
Arithmetic pipeline design
Unsolved Papers 2011 - 2006

Advanced Computer Arc.

Chapter 1

Parallel Computer Models

Q.1.

What is multiprocessors? What are the types of multi-processors?

Ans. A multiprocessor structure is an interconnection of two or more than two

CPUs with memory as input-output apparatus. Multiprocessors are
grouped as multiple instruction stream, multiple data stream (MIMD)
systems.
some similarities are found between multiprocessor & multicomputer
organization since both support simultaneous operations. However there is
an important peculiarity between a system with multiple computers & a
system with multiple processors. Computers are interconnected with each
other with communication lines to make a computer network. The network
comprise of numerous autonomous computers that may or may not converse
with each other. A multiprocessor system is governed by one operating
system that present inter-connection between processors & all the units of the
system assist in the solution of a problem very large scale integrated circuit
technology has abridged the cost of computer to such a low level that the
concept of applying multiple processors to meet system performance needs
has turn out to be an smart design prospect.
Multiprocessing develop the trustworthiness of system so that a failure or
error in one part has narrowed impact on rest of system. If a fault roots one
processor to fail, a second processor can be allocated to complete the
functions of disabled processor. The entire system can keep on functioning
suitably with possibly some loss in efficiency. The benefits resulting from a
multiprocessor organisation is better system performance. The system derives
its high performance from the fact that manipulation can advance in parallel
in one of two ways:
1. Multiple independent tasks can be completed to work in parallel.
2. A single task can be divided into multiple parallel tasks.
Example 1.
a computer system where are processor performs the
computations for an industrial process control while others monitors control
various parameter such as temperature and flow rate.

Example 2. a computer where are processor performs high speed floating

point mathematical computations and another take care of routine data
processing tasks.
Multiprocessing can improve performance by decomposing a
program into parallel executable tasks. This can be achieved in one of two
methods:
The user can explicitly declare that certain tasks of the program be executed
in parallel. This must be done prior to load the program by specifying the
parallel executable segments. Most multiprocessor constructs give an
operating system with programming language construct appropriate for
identifying parallel processing.
The other, more efficient way is to provide a compiler with multiprocessor
software that can automatically detect parallelism in a userss program. The
compiler checks for data dependency in the program. If a program depends
on data generated in another part, the part yielding the needed data must be
executed first. However two parts of a program that do not use data
generated by each can run concurrently. The parallelizing compiler checks
the complete program to detect any possible data dependence. These that
have no data dependency are then measured for concurrent scheduling on
different processors.
Multiprocessors are grouped according to their memory is organized. A
multiprocessor system with common shared memory is grouped as shared
memory ( tightly coupled) multiprocessor. This does not prohibit each
processor from having its own local memory. In fact, most commercial tightly
coupled multiprocessor offer a cache memory with each CPU. In addition
there is a global general memory that all CPUs can access. Information can
thus be shared amid the CPU by placing it in the general global memory.
An alternative model of microprocessor is the distributed memory or loosely
coupled system. Each processor element in a loosely coupled system has its
own private local memory. The processors are tied together by a switching
scheme designed to route information from one processor to another through
a message passing scheme. The processors relay program is data to other
processors in packets. A packet consists of an address, the data content and
some error detection code. The packets are addressed to a specific processor
or taken by first available processor, depending on the communication
system used. Loosely coupled systems are most efficient when the interaction
between tasks is minimal, whereas tightly coupled systems can tolerate a
higher degree of interaction between tasks.

Advanced Computer Arc.

Q.2. What is parallel processing?

Ans. Parallel processing refer to a huge class of method that are used to offer
simultaneous data processing jobs for increasing the computational speed of
system. as an alternative of processing each instruction sequentially as in
conventional computer, a parallel processing system is able to carry out
concurrent data processing to attain faster execution time. The idea of parallel
processing is to speed up the computer processing capability also increase its
throughput, i.e., the amount of processing that can be done during an interval
of time. Parallel processing at higher level of complexity can be realized by
including multiplicity of functional units that do identical or different
operations concurrently. Parallel processing is recognized by giving out the
data among the multiple functional units. For example the arithmetic logic
and shift operations
can be alienated into three units and the operands diverted to each unit under
the supervision of control unit.
Singe Instruction stream Single Data Stream (SISD)
IS

CU

IS

PU

DS

Single Instruction Multiple Data Stream (SIMD)

MM

10

PU1

PU2

DS1

DS2

MM1

MM2

CU

PUn

DSn

MMn

IS

CU: Control Unit

PU: Processing Unit
MM: Memory Module
These are divers of ways that parallel processing can be classified. One
classification introduced by M.J. Flynn considers the organization of
computer system by number of instructions and data items that are
manipulated simultaneously. The normal operation of a computer is to fetch
instructions from memory and execute them in the processor. The sequence
of instructions read from memory constitutes an instruction stream. The
operations performed on the data is processor constitutes a data stream
parallel processing may be occur in the instruction stream, in data stream or
both.
Single instruction stream, single Data stream (SISD)
Single instruction stream, multiple data stream (SIMD)
Multiple instruction stream, single data stream (MISD)
Multiple instruction stream, multiple data stream (MIMD)

Advanced Computer Arc.

IS1

CU1

IS2

CU2

IS1

IS2

11

PU1

PU2
MM1

MM2
ISn

IS3

CUn

ISn

MMn
IS2

IS1

DS

PUn

Multiple Instruction Stream Single Data Stream (MISD)

IS1

CU1
CU2

PU1
PU2

CUn

ISn

PUn

MM1

DS1
DS2

DSn

MM2

MMn

ISn
IS2
IS1

Multiple Instream stream Multiple Data Stream (MIMD)

Q.3.

Explain the state of computing?

Ans. Modern computers are equipped with powerful hardware facilitates driven
by extensive software packages. To asses the state of computing we first
review historical milestones in the development of computers.

12

Computer Generations
Over the past five decades, electronic computers have gone through fine
generations of development. Each of first three generations lasted about 10
years. The fourth generations covered a time span of 15 years. We have just
entered the fifth generations with the use of processors & memory devices
with more than 1 million transistors on solo silicon chip. The table shows the
new hardware and software features introduced with each generation. Most
features introduced in earlier generations have been passed to later
generations.
Five Generations of Electronic Computers
Generation
First
(1945-54)
Second
(1955-64)
Third
(1965-74)
Fourth
(1975-90)
Fifth
(1991
present)

Technology & Architecture

Vaccuum tubes & relay
memories, CPU motivated
by Pc
& accumulator, fixed point
arithmetic.
Discrete transistors and core
memories, floating point
arithmetic, I/O processors,
multiplexed memory access.
Integrated circuits (SSI-MSI),
microprogramming,
pipelining, cache &
lookahead processors.
LSI/VLSI & semi conductor
memory, multiprocessors,
vector supercomputers, multi
computers.
ULSI/VHSIC processors,
memory & switches, high
density packaging, scalable

Software & Application

Machine/assembly
languages, single user, no
subroutine linkage,
programmed I/O using
CPU.
HLL used with compilere,
subroutine libraries, batch
processing monitor.
Multiprogramming & time
sharing OS, multi user
applications.
Multiprocessor OS,
languages, compilers &
environment for parallel
processing.
Massively parallel
processing, grand challenge
applications, heterogenous
processing.

Representative
System
ENIAC, Princeton,
IAS, IBM 701

IBM 7090, CDC

1604, Univac
LARC.
IBM 360/370, CDC
6600, TI- ASC,
PDP-8
VAX 9000, Gay
XMP,
IBM 3090 BBN
TC 2000
Fujitsu VPP 500,
Gay/MPP,
TMC/CM-5, Intel
paragon.

architectures.

In other words, the latest generation computers have inherited all the bad
ones found in previous generations.

Advanced Computer Arc.

Q.4.

How is computer Architecture developed?

Ans. Over the past four decades, computer architecture has gone through
evolutional rather than revolution changes sustaining features are those that
were proven performance delivers.
According to the figure we started with the Von Neumann architecture
built as a sequential machine executing scalar data. Sequential computers
improved from bit serial to word-parallel operations & from fixed point to
floating point operations. The Von Neumann architecture is slow due to
sequential execution of instructions in programme.
Lookahead, Paralleism and Pipelining : Lookahead techniques were begin to
prefetch instructions in order to overlap I/E (instruction fetch/decod e and
execution) operations and to enable functional parallelism. Functional
parallelism was supported by two approaches: One is to use multiple
functional units simultaneously and the other is to practice pipelining at
various processing levels.
The latter embrace pipelined instruction execution, pipelined arithmetic
computations and memory access operations. Pipelining has proven
especially attractive in performing identical operations repeatedly over vector
data strings. Vectors operations were originally carried out implicitly by
software controlled looping using scalar pipeline processors.
Flynns Classification: Michael Flynn (1972) established nomenclature of a
variety of computer architectures based on concept of instruction and data
streams. Traditional sequential machines are SISD (single instruction stream
over
a
single
data
stream).
Vector
computers
are
set
with scalar and vector hardware or emerge as SIMD (single instruction
stream over multiple data streams). Parallel computers are called MIMD
(multiple instruction streams over multiple data streams) machines.
Q.5.

What is Parallelism? What are the various conditions of parallelism

Ans. Parallelism is the major concept used in today computer use of multiple
functional units is a form of parallelism within the CPU. In early computer
only one arithmetic & functional units are there so it cause only one operation
to execute at a time. So ALU function can be distributed to multiple
functional units, which are operating in parallel.
H.T. Kung has recognized that there is a need to move in three areas namely
computation model for parallel computing, inter process communication in
parallel architecture & system integration for incorporating parallel systems
into general computing environment.

13

14

Conditions of Parallelism :
1. Data and resource dependencies : A program is made up of several part,
so the ability of executing several program segment in parallel requires that
each segment should be independent other segment. Dependencies in various
segment of a program may be in various form like resource dependency,
control depending & data depending. Dependence graph is used to describe
the relation. Program statements are represented by nodes and the directed
edge with different labels shows the ordered relation among the statements.
After analyzing dependence graph, it can be shown that where opportunity
exist for parallelization & vectorization.
Data Dependencies: Relation between statements is shown by data
dependences. There are 5 types of data dependencies given below:
(a) Antidependency: A statement S2 is antidependent on statement ST1 if ST2
follows ST1 in order and if the output of ST2 overlap the input to ST1 .
(b) Input dependence: Read & write are input statement input dependence
occur not because of same variables involved put because of same file is
referenced by both input statements.
(c) Unknown dependence: The dependence relation between two statement
cannot be found in following situation
The subscript of variable is itself subscribed.
The subscript does not have the loop index variable.
Subscript is non linear in the loop index variable.
(d) Output dependence: Two statements are output dependence if they
produce the same output variable.
(e) Flow dependence: The statement ST2 is flow dependent if an statement
ST1 , if an expression path exists from ST1 to ST2 and at least are output of ST,
feeds in an input to ST2 .
2. Bernsteins condition : Bernstein discovered a set of conditions depending
on which two process can execute in parallel. A process is a program that is in
execution. Process is an active entity. Actually it is an stumbling block of a
program fragment defined at various processing levels. I i is the input set of
process Pi which is set of all input variables needed to execute the process
similarly the output set of consist of all output variable generated after
execution of all process P i. Input variables are actually the operands which
are fetched from the memory or registers. Output variables are the result to
be stored in working registers or memory locations.

Advanced Computer Arc.

Let there are 2 processes P1 & P2

Input sets are I1 & I2
Output sets are O1 & O2
The two processes P 1 & P2 can execute in parallel & are directed by P 1 /P2 if &
only if they are independent and do not create confusing results.
3. Software Parallelism : Software dependency is defined by control and data
dependency of programs. Degree of parallelism is revealed in the program
profile or in program flow graph. Software parallelism is a function of
algorithm, programming style and compiler optimization. Program flow
graphs shows the pattern of simultaneously executable operation. Parallelism
in a program varies during the execution period.
4. Hardware Parallelism : Hardware Parallelism is defined by hardware
multiplicity & machine hardware. It is a function of cost & performance trade
off. It present the resource utilization patterns of simultaneously executable
operations. It also indicates the performance of the processor resources.
One method of identifying parallelism in hardware is by means by number of
instructions issued per machine cycle.
Q.6.

What are the different levels of parallelism :

Ans. Levels of parallelism are described below:

1. Instruction Level : At instruction level, a grain is consist of less than 20
instruction called fine grain. Fine grain parallelism at this level may range
from two thousands depending an individual program single instruction
stream parallelism is greater than two but the average parallelism at
instruction level is around fine rarely exceeding seven in ordinary program.
For scientific applications average parallel is in the range of 500 to 300 fortran
statements executing concurrently in an idealized environment.
2. Loop Level : It embrace iterative loop operations. A loop may contain less
than 500 instructions. Some loop independent operation can be vectorized for
pipelined execution or for look step execution of SIMD machines. Loop level
parallelism is the most optimized program construct to execute on a parallel
or vector computer. But recursive loops are different to parallelize. Vector
processing is mostly exploited at the loop level by vectorizing compiler.
3. Procedural Level : It communicate to medium grain size at the task,
procedure, subroutine levels. Grain at this level has less than 2000
instructions. Detection of parallelism at this level is much more difficult than
a finer grain level. Communication obligation is much less as compared with

15

16

that MIMD execution mode. But here major efforts are requisite by the
programmer to reorganize a program at this level.
4. Subprogram Level : Subprogram level communicate to job steps and
related subprograms. Grain size here have less than 1000 instructions. Job
steps can overlap across diverse jobs. Multiprogramming an uniprocessor or
multiprocessor is conducted at this level.
5. Job Level : It corresponds to parallel executions of independent tasks on
parallel computer. Grain size here can be tens of thousands of instructions. It
is handled by program loader and by operating system. Time sharing & space
sharing multiprocessors explores this level of parallelism.
Q.7.

Explain Vector super computers?

Ans. Program & data are first loaded into the main memory from a host computer.
All instructions are first decoded by the scalar control unit. If the decoded
instruction is a scalar operation or program control operation it will be
directly executed by scalar processor using the scalar functional pipelines.
If the instruction is decoded as a vector procedure, it will be sent to the vector
control unit. This control unit will supervise the flow of vector data amid the
main memory & vector functional pipelines. The vector data flow is
synchronized by control unit. A number of vector functional pipelines may be
built into a vector processor.
Computers with vector processing capabilities are in demand in specialized
applications. The following are symbolized application areas where vector
processing is of utmost importance.
Long Range weather forecasting
Petroleum explorations
Medical diagnosis
Space flight simulations

Advanced Computer Arc.

17

Vector Processor Models

Scalar Processor
Scalar
Functional
Pipelines
Vector processor

Scalar Instructions

Vector
control
unit

Scalar
control
unit

Control
Instructions

Main
menory

Scalar
data

Mass
storage

Vector
func. pipe
Vector
registers

Host
computer

Vector
function pipe

The Architecture of vector super computer

Q.8.

What are the different shared memory multiprocessor models?

Ans. The most popular parallel computers are those that execute programs in
MIMD mode. There are two major classes of parallel computers: shared
memory multiprocessor & message passing multi computers. The major
distinction between multiprocessors & multicomputer lies in memory sharing
and the mechanisms used for interprocessor communication. The processor
in multiprocessor system communicate with each other through shared
variable in a common memory. Each computer node in a multicomputer
system has a local memory, unshared with other nodes. Inter process
communication is done through message passing among nodes.
There are three shared memory multiprocessor models:1. Uniform memory access (UMA) model
2. Non-uniform memory access (NUMA) model
3. Cache only memory Architecture (COMA) model
These models are differ in how the memory & peripheral resources are
shared or distributed.

18

1. UMA Model:

The UMA multiprocessor model

P1

P2

Processor

Pn

System Interconnect
(Bus, Crossbar, Multistage network)

I/O

SM1

Shard manery

SMm

In this model the physical memory is uniformly shared by all the processors.
All processors have equal access time to all memory words, which is why it is
called uniform memory access. Each processor may use a private cache.
Peripherals are also shared.
Multiprocessors are called tightly coupled systems for its high degree of
resource sharing.
UMA model is suitable for time sharing applications by multiple users. It can
be used to speed up the execution of single large program in time critical
application.When all processors have equal access to all peripheral devices,
the system is called a symmetric multiprocessor. In this case, all the
processors are equally capable of running programme, such as kernel.
In an asymmetric multiprocessor, only one or subset of processors are
executive capable. An executive or master processor can execute the
operating system and handle I/O. The remaining processors called attached
processors (AP) runs user code under the supervision of master processor.

Advanced Computer Arc.

19

2. NUMA model: A NUMA multiprocessor is a shared memory system in

which the access time diverge with the location of memory word.
Two NUMA machine models are depicted. The shared memory is physically
distributed to all processors, called local memories. The collection of all local
memories forms a global address space accessible by all processors.
It is quicker to access a local memory with a local processor. The access of
remote memory attached to other processors takes longer due to the added
delay through the interconnection network.

LM1

P1

LM2

P2

LMn

Pn

Inter
Connection
Network

Shared Local Memories

In the hierarchial cluster Model processors are divided into several clusters.
Each cluster may be UMA or NUMA Each cluster is connected to shared
memory modules. All processors of a single cluster uniformally access the
cluster shared memory modules. All cluster equally access to global memory
access time to cluster memory is shorter then that of global memory.

20

GSM

GSM

GSM

Global Interconnect Network

P1

P2

Pn

CSM

P1

CSM

P2

CSM

Pn

CSM

CSM

CSM

A hierarchical cluster models

3. Cache Only Memory Architecture:
This model is a special case of NUMA machine where distributed main
memories are replaced with cache memory. At individual processor node,
there is no memory chain of command(hierarchy). All cache made a global
address space. Depending on interconnection network used, directories may
be
used to help in locating copies of cache blocks example of COMA i ncludes
Swedish Institute of Computer Sciences Data Diffusion machine (DDM).
Interconnection Network

Advanced Computer Arc.

Q.9.

21

What is Distributed Memory Multicomputers?

Ans. A system made up of multiple computers( nodes), interconnected by a

message passing network. Each node is autonomous computer consisting of a
processor, local memory and from time to time attached disks or I/O
peripherals.

Message passing interconnection network

(mesh, ring etc.)

All local memories are private & accessible only by local processors. This
network presents point-to-point static connection among nodes. Inter node
communication is carried out by passing messages through the static
connection network.
Q.10. Explain SIMD Computers?
Ans. SIMD connotes single instruction stream and multiple data stream. These
computers are array processors. There are multiple processing elements
which are supervised under same control unit. Each processing element
receives same instruction but operate on different data from distinct module,
SIMD Machine Model
An operational model of an SIMD computer is specified by 5- Triple.
M = [ N, C, I, M, R]
Where
(1) N is the number of processing elements (PEs) in the machine.
(2) C is the set of instructions directly executed by control unit including
scalar and program flow control instructions.

22

(3) I is set of instructions broadcast by CPU to all PEs for parallel execution.
These include arithmetic, logic, data routing, masking and other local
operations executed by each active PE over data within that PE.
(4) M is the set of masking schemes, where each mask partitions the set of PEs
into enabled & disabled subsets.
(5) R is the set of data routing functions, specifying various patterns to be set
up in the inter connection network for inter PE communications.
Q.11. What are the Architectural development tracks?
Ans. Architecture of todays systems pursue development tracks. There are mainly
3 tracks. These tracks are illustrious by likeness in computational model &
technological bases.
1. Multiple Processor tracks: multiple processor system can be shared
memory or distributed memory.

(a) Shared Memory track:

Standard/Dash
Itlinosis
cedar

Fujitsu VPP 500

KSR 1

CMU/
C.mmP
NYU/
Ultra Computer

IBM RP

BBN Butterfly

Fig. Shared Memory track

It shows track of multiprocessor development employing a single address
space in the entire system c. mmp was a UMA multiprocessor. The c.mmp
project poincered shared memory multiprocessor development not only in
the cross architecture but in multiprocessor operating system development.
Illinois Codar project and NYO ultra computer project both were developed
with a single address space. Both use multi stage network as system inter
connect.
Standard Dash is a NUMA multiprocessor with distributed memory forming
a global address space cache coherence is there with distributed directories.

Advanced Computer Arc.

23

KSR-1 is a COMA model.

Fujitsu UPP 500 is processor system with a cross bar inter connected shared
memories are distributed to all processor nodes.

(b) Message Passing track:

CUBE 2/6400
Cosmic Cuben

Inter iPscs
Mosaic

Intel paragon
MIT/J Machine

(2) Multivector & SIMD tracks

Multivector track
CDC Cyber 205- ETA 10

CDC 7600
Cray Y- mp

Cray/m PP

Cray 1
Fujitru, NEC, Hitachi Mode

The CDC 7600 was first vector dual processor system. There are 2 subtracks
derived from CDC-7600. The latest cray/mpp is a massively parallel system
with distributed shared memory.

24

(b) SIMD track

DAP 610
Goodyear MPP
CM5
Illiac IV
Mas Par MP1
BSP
IBM GF/11

3. Multi threaded and Dataflow tracks:

In case of multi threaded system, each processor can execute multiple context
at the same time. So multiple threading means there are multiple threads of
control in each processor. So multi threading hides long latency in
constructing large scale multiprocessors. This track has been tried out in
laboratories.
Multi threaded track

Tera
CDC 600

HFP
MIT/Alenrife

Data Flow Track

MIT tagged token
Static Data flow
Mamchester

Q.12. What are the two configurations of SIMD array processor.

Ans. Synchronous array of parallel processors is called array processor, which
made up of multiple processing element (PES).
SIMD array processor have 2 arrangement

Advanced Computer Arc.

25

arrangement I

I/O
Data & Instruction
Data bus

PE

CU
Memory
CU

PE

PEM

PEN

PEM

Interconnection Network

PEM

Control

arrangement I (Illiac IV)

First SIMD array configuration which are introduced in Illiac IV
computer. This is having N synchronized PEs, all are under the control of one
CU. Each PE is an arithmetic logic unit (ALU) with attached working register
and local memory PEM for storage.
Control unit has its own memory for storage of programs. First user
programs are loaded into CU memory and then CU decode all the
instructions and find out where the decade instructions should be executed.
Scalar and control instructions are executed inside CU and vector instructions
are broadcast to the PE for the distributed execution.

26

arrangement II
I/O
CU Memory

Control

CU

PE0

PE1

PEn-1

Alignment Network

M0

M1

Mn-1

arrangement II (BSP)
Main differences in arrangement I and II is in 2 aspects. First the local
memories are attached to the PEs are replaced by parallel memory module
shared by all the PEs through an alignment network. Second, inter PE
network is replace by the inter PE memory alignment network, which is
controlled by CU.
Example of configuration II is Burrough Scientific processor (BSP). There are
N PEs and P memory modules in configuration II. These two numbers (N and
P) are not equal and are relatively prime. The alignment network is a path
switching network between PEs and parallel memories.

Advanced Computer Arc.

27

Chapter 2

Program Partitioning or Scheduling

Q.1.

What are program flow mechanisms?

Ans. Traditional computers are founded on control flow mechanism by which the
order of program execution is explicitly stated in the user program. Data flow
computers have high degree of parallelism at the fine grain instruction level
reduction computers are based on demand driven method which commence
operation based on the demand for its result by other computations.
Data flow & control flow computers : There are mainly two sort of
computers. Data flow computers are connectional computer based on Von
Neumamm machine. It carry out instructions under program flow control
whereas control flow computer, executes instructions under availability of
data.
Control flow Computers : Control Flow computers employ shared memory
to hold program instructions and data objects. Variables in shared memory
are updated by many instructions. The execution of one instruction may
produce side effects on other instructions since memory is shared. In many
cases, the side effects prevent parallel processing from taking place. In fact, a
uniprocessor computer is inherently sequential due to use of control driven
mechanism.
Data Flow Computers : In data flow computer, the running of an instruction
is determined by data availability instead of being guided by program
counter. In theory any instruction should be ready for execution whenever
operands become available. The instructions in data driven program are not
ordered in any way. Instead of being stored in shared memory, data are
directly held inside instructions. Computational results are passed directly
between instructions. The data generated by instruction will be duplicated
into many copies and forwarded directly to all needy instructions.
This data driven scheme requires no shared memory, no program counter
and no control sequencer. However it requires special method to detect data
availability, to match data tokens with needy instructions and to enable the
chain reaction of asynchronous instructions execution.

28

Q.2.

Explain data flow architecture?

Ans. There are more than a few new data flow computer projects. Arvind and his
associates at MIT have developed a tagged token architecture for constructing
data flow computers.
Global path

nxn Routing Network

PE'

PE2

PEn

the global architecture

The global architecture incorporate n processing elements (PEs) inter

connected by an n x n routing network. The complete system supports
pipelined data flow operations in all n PEs. Inter PE communications are
done through the pipelined routing network.
Within each PE, the machine offer a low level token matching mechanism
which sends off only those instructions whose input data are already
available. Each datum is tagged with the address of instruction to which it
belongs and context in which the instruction is being executed. Instructions
are stored in program memory. Tagged tokens enter the PE through a local
path. The tokens can also be passed to the other PE through the routing
network. All internal circulation operations are pipelined without blocking.

Advanced Computer Arc.

29

From Routing Network

Local Path

Token
Match

Program
memory

1-Structure

Compute
Tag
ALU

Form Token

Interior Design of a Processing Element

You can imagine instruction address in a dataflow computer as replace the
program counter & the context identifier replacing the frame base register in
control flow computer. It is the machine job to match up data with same tag
to needy instructions. In so doing, new data will be produced with a new tag
indicating the successor instructions. Thus each instruction represents a
synchronization operation. New tokens are formed and circulated along the
PE pipeline for sense or to other PEs through global path, which is also
pipelined.
Q.3.

Explain Grain Sizes and Latency.

Ans. Grain Size or granularity is the amount of computation and manipulation

involved in a software process. The simplest way is to count the number of
instructions in a given (program segment). Grain size decides the basic
program segment chosen for parallel processing. Grain sizes are usually
explained as fine, medium or coarse, depending on the processing levels
involved.
Latency is a time of the communication overhead acquire amid subsystems
for example the memory latency is the tune required by processor to access

30

the memory. The time required for two processes to synchronize with each
other is called synchronization latency; computational granularity and
communication latency are closely related.
Q.4.

How can we partition a program into parallel branches, program modules,

microtasks or grains to yield the shortest possible execution time?

Ans. There exists a tradeoff among parallelism and scheduling overheads. The
time complexity entail both computation and communication overheads. The
program partitioning entail the algorithm designer, programmer, compiler,
operating system support etc.
The concept of grain packing is to apply five grain first in order to achieve a
higher degree of parallelism. Then one combines multiple fine grain nodes
into a coarse grain node if it can remove redundant communications delays
or lessen the overall scheduling overhead.
Usually, all five grain operations within a single coarse, grain node are given
to some processor for execution. Fine grain partition of a program often
demands more inter processor communication than that required in a coarse
grain partition. Thus grain pickings offers a tradeoff between parallelism and
scheduling. Internal delays amid fine grain operations within the same coarse
grain node are negligible because the communication delay is given chiefly
by inter processor delays rather than by delays within the same processor.
The selection of optimal grain size is meant to get the shortest schedule for
the nodes on a parallel system.
ppp

Advanced Computer Arc.

31

Chapter 3

System Interconnect Architecture

Q.1.

Explain different network properties?

Ans. The topology of an interconnection network can be either static or dynamic.

Static networks are created point-to-point direct connections which will not
alter during execution. Dynamic networks are applied with switched
channels, which are dynamically configured to match the communication
demand in user programs.
Static networks are used for fixed connections amid sub systems of a
centralized system or multiple computing nodes of a distributed system.
Dynamic networks consist of buses, crossbar switches, multistage networks,
which are often used in shared memory multi processors. Both types of
network have also been employ for inter PE data routing in SIMD computers.
In general, a network is characterized by graph of finite number of nodes
linked by directed or undirected edges. The number of nodes in the graph is
called the network size.
Node Degree and Network Diameter : The number of edges incident on a
node is called the node degree d. In the case of unidirectional channels, the
number of channels into a node is the in; degree and that out of a node is the
out degree. Then the node degree is the total of the two. The node degree
reveals the number of I/O ports required per node and their cost of a node.
Hence, the node degree should be kept a constant, as small as possible in
order to reduce cost. A constant node degree is very much preferred to get
modularity in building blocks for scalable systems.
The diameter D of a network is the maximum shortest path amid any two
nodes. The path length is measured by the number of links visited. The
network diameter show the maximum number of distinct hops amid any two
nodes, thus giving a figure of communication pros for the network. thus, the
network diameter should be as small as doable from communication point of
view.
Q.2.

What is Bisection Width?

Ans. When a specified network is cut into two identical halves, the minimum
number of edges along the cut is termed as channel bisection width b. In the

32

case of communication network, each edge match up to a channel with w bit

wires. Then the wire bisection width is B = bw. This parameter B reflects the
wiring density of a network. When B is fixed, the channel width w = B/b.
Thus the bisection width offer a fine estimate of maximum communication
band width along the bisect ion of a network. Rest cross sections should be
bounded by bisection width.
Q.3.

What is Data routing functions? Describe some data routing functions?

Ans. Data routing networks is used for inter PE data exchange. Data routing
network can be static or dynamic. In multicomputer network data routing is
achieved by message among multiple computer nodes. Routing network
reduces the time required for data exchange and thus system performance is
enhanced. Commonly used data routing functions are shifting, rotation,
permutations, broadcast, multicast, personalized communication, shuffle etc.
Some Data routing functions are described below:
(a) Permutations: Let there are n objects, and then there are nf permutations
by which n objects can be recorded. Set of all permutations form a
permutation group with respect to composition operation. Generally cycle
notation is used to specify permutation function. Cross can be used to
implement the permutation. Multi stage network can implement some of the
permutations in one or multiple passes through the network. Shifting and
broadcast operation are also used to implement permutation operation.
Permutation capability of a network is used to indicate the data routing
capacity. Permutation speed dominates the performance of data routing
network, when n is large.
(b) Hypercube routing function: Three dimensional cube is shown below:
Routing functions are defined by three bits in the node address. Bit order is
C2 C1 Co. Data can be exchanged among adjacent nodes which differs in the
least significant bit Co as shown below.

000

001

010

011

100

101

110

Routing by least significant bit, Co

111

Advanced Computer Arc.

33

Similarly routing pattern by using bit C1 & C2 is shown below:

000

001

010

011

100

101

110

111

Routing by middle bit, C 1

000

001

010

011

100

101

110

111

Routing by most significant bit, C 2

common pattern informs that n-dimensional, cube has n-routing functions,
which are defined by each bit of the n-bit address. These data
routing task are used in routing messages in a hypercube multi workstation.
(c) Broad cast & Multicast: Broad cast is one to all mapping. This is achieved
by SIMD computers using a broadcast bus extending from array controller to
all PEs. A mechanism is used to broadcast a message in message passing
multi computer. Multicast means mapping from one subset to another. There
is a variation of broadcast called personalized broadcast. Personal ized
broadcast sends messages to only selected receivers. Broadcast is a global
operation in multi computer. Personalized broadcast may have to be
implemented with matching of destination codes in the network.
Q.4.

What are static interconnection networks?

Ans. Static interconnection networks have many topologies. These topologies are
classified according to the dimensions required for layout. Example one
dimension, 2 dimensional, 3- dimensional. One dimensional incorporated
linear array which is used for a number of pipelined architecture 2dimensional includes topology: ring, star, mesh and systolic array. 3

34

dimensional embrace completely connected chordal ring, 3-cube and 3cube connected cycle network.
One dimensional topology
Linear Array: In this N-nodes are connected by N-1 links in the
line.
In linear array, each internal node has degree equal to 2. Each external node
or terminal nodes have degree equal to 1. Diameter is N -1, which is long for
large value of N. The bisection width is b = 1. Linear arrays are actually
simplest connection topology. arrangement of linear assays is not symmetric.
So when N is large, then communication inefficiency is there.
0
0

1
0

2
0

3
0

7
0

6
0

5
0

4
0

8
0

9
0

10
0

11
0

15
0

14
0

13
0

12
0

For small N, say N = 2, it is economic to implement a linear array. As

diameter increases linearly with respect to N, it should not be used for large
N. Linear array is different from bus which is time shared through switching
among many nodes attached to it. Concurrent use of different sections of the
structure of different source and destination pairs is permitted in linear
arrays.
Two Dimensional topology
Star: Star is a two level tree with a high node degree of = N-1 and a small
constant diameter of 2. Star topology is used in systems with a centralized
supervisor node.

Advanced Computer Arc.

35

O
O

10

11
O

12
O

0
O

O1

9O
O

O2

8O
O3

O
7

O
6

Star topology

O
4
O
5
Ring topology

Systolic array: For implementing fixed algorithm this type of pipelined array
architecture
is
used.
Figure
of systolic array
topology is
shown:
O
O

O
O

O
O

O
O

It is designed for matrix multiplication. Degree of interior nodes are equal to

6. Systolic array are pipelined which multi dimensional flow of data streams.
Systolic array matches the communication structure of the algorithm. It is
similar to the fixed inter connection and synchronous ratio over special
applications like signal/image processing. It is difficult to program and thus
having limited applications.
Three Dimensional Topology
3- Cube
A 3- cube having B-nodes is shown below A

36

O
O

O
O

O
O

O
O

3- Cube
A 4- cube is made by interconnecting the corresponding nodes of
two 3-cubes. Node degree of n-cube is equal to n and so does the
network diameter. Nodes degree increases linearly with respect to the
dimensions, thus making hypercube difficult to use as a scalable architecture.
Cube-Connected Cycles
Improved architecture of hypercube is the cube connected cycles. 3- Cubes
are customized to form 3-cube connected cycles (CCC).
Thus, K-cube connected cycle can be made from the L-cube with n = 2x cycles
nodes. Each vertex of K-dimensional hyper cube is replaced by a ring of Knodes thus a k- cube is translated into a K-CCC with K x 2K nodes.
Q.5.

Discuss various Dynamic Connection Network.

Ans. For multipurpose and general purpose application we always use dynamic
connections. Dynamic connection implements all common wrication patterns
based on program demands. Fixed connections with switches on arbiters are
used along the connecting path to provide the dynamic connectivity.
Basically there are 2 classes of dynamic interconnection network-single stage
and multistage network.
Single Stage Network: Single stage network is a switching network with Ninput selectors (Is) and N output selectors (Os). Each input selector is 1- to D
demultiplexer and each output selector is an m to -1 multiplexer where 1 =
D = N and 1 = m = N. Single stage network is called recirculating network.
Data items are recirculate through the single stage determines the number of
recirculations required. Crossbar switching network is a single stage network
with D = M = N. For establishing the desired path different control signals are
applied to all input selectors and output selectors.

Advanced Computer Arc.

37

Multistage Network: Many stages of interconnected switchers forms a

multistage MIMD network. Multistage network are characterized by 3
properties
(a) Switch box

IS
O

IS
O

IS
1

IS
1

IS
N-1

N-1

N-1

IS
N-1

(b) Network topology

(c) Control structure
Multistage network uses many switch boxes. Each switch is an interchange
device with 2 inputs & 2-outputs. There are 4- stage of switch box straight,
exchange, upper broadcast and lower broadcast. A two function switch box
can be either in switch can be in any of four legitimate states. Switching box
and their inter connection stages are given on subsequent
page.
A multistage network connects an arbitrary input terminal to an arbitrary
output terminal. There are two kind of multi stage network: one sided or two
sided. One sided network is called fill switches and they include input and
output ports on the same size. Two-sided multistage network have an input
side and output side divided into classes:
Blocking , rearrange able and blocking.

38

a0
a1

b0
Switch box

b1

a0

>

b0

a1

>

b1

a0

b0

a1

b1

a0

b0

a1

b1

a0

b0

a1

b1

A two by two switching box and its four interconnection states

ppp

Advanced Computer Arc.

39

Chapter 4

Processors and Memory Hierarchy

Q.1.

What is difference between RISC and CISC ?

Ans.
Properties

CISC

RISC

1. Number of instruction

It differs from 120-350

It is less than 100

Variable instruction and

Fixed format instruction/

data format is used.

data is used.

differ from 12-24

differ from 3 to 5

Vary from 8 to 24

Vary from 32-192

Large number of memory

Less number of memory

reference instructions.

reference instruction are

used. Only load and store

used in set architecture

2. Instruction/data format
used
3. Number of addressing
modes
4. Number of general
purpose registers used
5. Number of memory
reference instruction used
are memory reference
instruction.
6. Number of memory
Large number of memory Less number of memory
reference instruction used
reference instructions. reference instruction are
used. Only load and
store
are memory
instructions.
7. High level language
instructions

Directly employed in

Not directly employed in

hardware

hardware.

Execution competence

Execution competence is not

increased.

increased.

Micro programmed

Hardwired control unit is

control unit is used.

used.

implementation
8. Execution efficiency
9. Control logic used

40

10. Use of control memory

Control memory is employed. No use of control memory.

11. Cache memory

Unified cache memory

Splited cache is employed.

is employed.
12. Clock rate

Q.2.

35-50MHz

50-150 MHz

What is Super Scalar Processors?

Ans. In a super scalar processor, multiple instructions are employed, this means
that multiple instructions are issued per cycle and multiple results are created
per cycle.
In simple scalar processor one instruction executes per cycle. Only one
instruction is employed per cycle and only one completion of instruction is
probable per cycle.
Super Scalar process are intended to utilize more instruction level parallelism.
Super scalar operate basically in parallel.
Super Scalar Architecture
It involves highly multipored register files. Their input parts are required for
each EU.
Super Scalar processors acknowledge a usual sequential stream of
instructions but can generate more than are instructions to the EUs in each
cycle. Super scalar processors do not presume dependency free code. They
cope with dependencies themselves using hardware. Super scalar processors
with the same degree of parallel execution are considerable more
multifaceted.

Advanced Computer Arc.

Cache
mimory

41

Fetch
unit

Decode
unit
Multiple instruction

EU

Instruction/Data
Data
EU : Execution unit

Q.3.

EU

EU

Register file

What is VLIW architecture?

Ans. The VLIW architecture is generalized. It includes well-known concepts:

1. Horizontal Microcoding
2. Super scalar processing
A typical VLIW architecture machine has instruction words of hundreds of
bits in length VLIW stands for very long instructions word.

42

Main
memory

Register file

Load/
store
unit

F.P.
add
unit

Integer
ALU

Branch
unit

VLIW processor
VLIW processor expect dependency free code i.e. multi operation code. VLIW
processors are statically programmed. VLIW notion is taken from horizontal
micro coding. Different fields of the long instruction word carry the opcodes
to be send off to different functional units.

Advanced Computer Arc.

Cache
memory

43

Fetch
unit
Single multi operation interuction

EU

EU

EU

EU

Register file
Multioperation
instruction

Instruction control
Data
Eu : Execution unit

Q.4. Explain Symbolic Processors?

Ans. Symbolic processing applied in numerous cases including theorem proving,
pattern recognition, expert systems, text retrieval, cognitive science and
machine intelligence, knowledge engineering. In three applications, data and
knowledge representation, primitive operations, I/O and special architectural
features are different them in numeral computing symbolic processors are
named prolog processors or symbolic manipulators.
Characteristics of Symbolic Processing
Attributes

Characteristics

1. Knowledge

Lists, relational databases, scripts, semantic nets, frames,

production system.

2. Common operation

Search, sort, pattern matching, filtering, unification, text

retrieval, reasoning

3. Memory Requirements

Large number with intensive access pattern. Addressing

is often content based.

4. Communication pattern

Message traffic varies in size and destination granularity

& format of message unit change with applications.

5. Properties of Algorithm

Non-deterministic, possibly parallel and distributed

computations.

44

6. Input-Output Requirements

User-guided programs, intelligent-person

interface, input can be graphical.

machine

7. Architecture features

Parallel update of large knowledge bases, dynamic load

balancing, dynamic memory allocation, hardware
supported garbage collection.

Q.5. What is Virtual memory? In how many classes virtual memory system is
categorized?
Ans

Virtual memory is a notion used in some large system, that permits user to
make program as through large memory space were present, equal to the
totality of auxiliary memory. In memory ladder programs and data are first
stored in auxiliary memory. Portion of program and data are then brought
into main memory as they are needed by CPU. Each address referenced by
CPU goes through an address mapping from virtual address to physical
address in memory.
Hence virtual memory forever give an false impression that they have large
memory at their disposal, even through computer has relatively small
memory.
Virtual memory based system provides a mechanism for translating program
generated address into correct main memory locations. This all process is
done dynamically, when process are executing in main memory. The
translation a mapping is handled by automatically using hardware of
mapping table. The address used by programs are called virtual address as
such addresses set is called address space. An address in main memory is
called physical address set of such address is called memory space.
Virtual Memory System is grouped in 2 classes:(1) Those with fixed sized blocks called pages.
(2) Those with variable size block called segments.
Paging: It is a memory management method that allows the physical address
space of a method to be non-continuous.

Advanced Computer Arc.

45

t
CPU

P d

f0000

Page table
Physical
memory

Fig. Paging Hardware

s
limit Base

CPU

Sigment table

s d

yes
L

Physical
memory

Fig. Segmentation hardware

Q.6. Explain Backplane Bus System?
Ans

A backplane bus interconnects processors, data storage and peripheral

devices in a tightly coupled hardware arrangement. The system bus must be
designed to permit communication amid devices on the bus without
disturbing the interval activities of all devices attached to the bus. Timing
protocols must be established to arbitrate among multiple requests.
Operational rules must be set to guarantee orderly data transfers on the bus.
Signal lines on the backplane are often functionally grouped into several
buses as shown in the figure. The four groups shown here are very similar to
those proposed in the 64 bit VME bus specification.

46

Various functional boards are plugged into slots on the backplane. Each slot
is given with one or more connectors for pop in the boards as shown by the
vertical arrows. For example one or two 96-pin connectors are used per slot
on the VME backplane.
CPU Board

Memory Board

Processor
and
Cache

Memory
Array

Functional
modules

Functional
Modules

System clock
driver, daisy
chain drives
Powder driver
Bus timer
Arbiter

Interface
Logic

Interface
Logic

Interface
Logic

Bus controller

Slot 1

Backplanes (signal lines & connectors)

Data transfer Bus (DT B)

DTB Arbitration Bus
Interrupt & Synchronization Bus
Utility Bus

Backplane buses, system interfaces with slot connections to various functional

boards in a multiprocessor system
ppp

Advanced Computer Arc.

47

Chapter 5

Pipelines Processors
Q.1.

What are the characteristics of Pipeline?

Ans.Pipelining refers to the temporal overlapping of processing pipelines. It
is more than assembly lines in computing that can be employed for
instruction processing. A basic pipeline process a sequence of tasks or
instruction, according to the following principle of operation.
Each task is subdivided into a number of consecutive tasks. The processing of
each single instruction can be broken down into four sub tasks:1. Instruction Fetch
2. Instruction Decode
3. Execute
4. Write back
It is supposed that there is a pipelined stage associated with each subtask.
The equal amount of time is available in each stage for performing the
required subtask.
All the pipeline stages work like an assembly line, that is, receiving their
input from the previous stage and delivering their output to next stage.
We also suppose, the basic pipeline operates clocked, in other words
synchronously. This means that each stage accepts a non input at start of
clock cycle, each stage has a single clock cycle available for performing the
required operation and each stage increases the result to the next stage by the
beginning of subsequent clock cycle.

48

Single task

Subtask 1 Subtask 2 Subtask 3 Subtask 4 Subtask 5 Subtask 6

i/p

Stage 1

Stage 2

Stage 3

O/P

Stage n

Q.2.

Explain Linear Pipeline Processors ?

Ans. A linear Pipeline processor is a flow of processing stages which are linearly
connected to perform a fixed function over a stream of data flowing from one
end to other. In modern computers, linear pipelines are applied for
instruction execution, arithmetic computation, memory access operations.
A linear pipeline processor is built with the processing stages. External inputs
are inputted into the pipeline at the first stage S1 . The processed results are
passed from stage Si to stage Si+1 for all i = 1,2.K-1. The final result
appear from the pipeline at the last stage Sk. Depending on the control of
data flow along the pipeline, linear pipelines are formed in two group.
Asynchronous Model: Data flow amid adjacent stages in asynchronous
pipeline is controlled by handshaking protocol. When stage S1 is ready to
transmit, it sends a ready signal to Si + 1. After stage Si+1 receives the
incoming data, it returns an acknowledge signal to Si.
Input
Ready
Ack

S1

Ready
Ack

S2

Ready
Ack

Onput
Sk

Ready
Ack

An Asynchronous pipeline Model

Synchronous Model: Clocked latches are used to interface between stages.
The latches are made with master slave flip flops, which can detach inputs
from outputs upon the arrival of a clock pulse, all latches transfer data to the
next stage simultaneously.

Advanced Computer Arc.

49

L
Output

S2

S1
o

Sk
o

Clock

A Synchronous pipeline Model

Q.3.

Explain Non linear Pipeline Processors?

Ans. A dynamic pipeline can be reconfigured to carry out variable functions at

different times. The traditional linear pipelines are static pipelines because
they are used to carry out fixed functions. A dynamic pipeline permit feed
forward and feedback connections besides the streamline connections. For
this reason, some authors call such a structures as non-linear pipeline.
Output
Input
S1

S2

Sk

Output

A three stage pipeline

This pipeline has three stages. Besides the streamline connections from S 1 to S2
and from S2 to S3 , there is feed forward connection from S2 to S3 and two
feedback associations from S3 to S2 and from S3 to S1 .
These feed forward and feedback connections make the scheduling of
consecutive event into the pipeline a non trivial task. With these connections,
the output of the pipeline is not necessarily from the last stage. In fact,
following different dataflow model, one can use the same pipeline to assess
different functions.

50

Q.4. What is Reservation Table in linear pipelining?

Ans. The utilization pattern of successive stages in a synchronous pipeline is
mentioned by reservation table. The table is essentially a space time diagram
depicting the precedence relationship in using the pipeline stages. For a K stage linear pipeline, K clock cycles are needed to flow through the pipeline.
1

Stages

S1

S2

S3

S4

Reservation table of 4-Stage

Q.5. What is Reservations table in Non-linear pipelining?
Ans. Reservation table for a dynamic pipeline become more complex and
interesting because a non-linear pattern is followed. For a given
non-linear pipeline configuration, multiple reservation tables can be
generated. Each reservation table will show evaluation of different function.
Each reservation table displays the time space flow of data through the
pipeline for one function evaluation. Different function may pursue different
paths on the reservation table.
1

Stages

S1

S2

S1

S2

S3

S1

S3

S1

8
X

Processing sequence
S2

6
X

S3

S1

Advanced Computer Arc.

51

Reservation table for function X

Q.6.

What is Instruction Pipeline Design?

Ans. A stream of instructions can be carry out by pipeline in an overlapped

manner. A typical instruction execution consists of a sequence of operations,
including
(1) Instruction fetch
(2) Decode
(3) Operand fetch
(4) Execute
(5) Write back phases
Pipeline instruction processing
A typical instruction pipeline has seven stages as depicted below in figures.

Fetch
F

Decode
D

Issue
I

Execute
E

Execute
E

Execute
E

Write
Back

Fetch stage (F) fetches instructions from a cache memory.

Decode stage (D) decode the instruction in order to find function to be
performed and identifies the resources needed.
Issue stage (I) reserves resources. Resources include GPRs, bases and
functional units.
The instructions are executed in one or several execute stages (E)
Write back stage (WB) is used to write results into the registers.
Memory lead and store (L/S) operations are treated as part of solution.
Floating point add and multiply operations take four execution clock
cycles.
In many RISC processors fewer cycles are needed.
Ideal cycles when instruction issues are blocked due to resource conflicts
before date Y and Z are located in.

52

the store of sum to memory location X must wait three cycles for the add
to finish due to flow dependence.
Q.7.

Explain Arithmetic Pipeline design?

Ans

Pipeline arithmetic units are typically set up in very high speed computers.
They are used to apply floating point operation, multiplication of fixed point
numbers and similar computations encountered in scientific problems.

The exponent are compared by subtracting them to determine their

difference.

a Exponents b

A Mentissa B

Segment 1 Compare exponents

by subtraction

Segment 2

Choose exponent

Allign mantissa

R
Segment 3

Segment 4

Add or subtract
Mantissa

Adjust exponent

Normalise Result

Advanced Computer Arc.

53

Arithmetic Pipeline for Addition & Subtraction

Exponent difference determine how many times the mantissa associated
with the smaller exponent must be shifted to the right.
This produces are alignment of two mantissas.
The teno mantissas are added or subtracted in segment 3.
Finally result is normalised in segment 4.
When a overflow occurs, the mantissa of the sum or difference is shifted
right and the exponent is incremented by one.
When an underflow occurs, the number of leading zeroes in the mantissa
determines number of left shifts in the mantissa and the number that must
be subtracted from the exponent.
ppp

54

Case Studies

Case Study 1.
Cache Policies.
Consider two alternate caches, each with 4 sectors holding 1 block per sector and
one 32-bit word
per block. One cache is direct mapped and the other is fully associative with LRU
replacement
policy. The machine is byte addressed on word boundaries and uses write allocation
with write
back.
1a) What would the overall miss ratio be for the following address stream on the
direct
mapped cache? Assume the cache starts out completely invalidated.
read 0x00 M
read 0x04 M
write 0x08 M
read 0x10 M
read 0x08 H
write 0x00 M Miss ratio = 5/6 = 0.8333

Case Study 2. Virtual Memory and Cache Organization.

The 742LX is a uniprocessor having up to a maximum of 64 MB of addressable

physical
memory. The cache, virtual memory, and TLB have the following attributes:
Cache
Virtual
Memory TLB
unified
virtual page size is 4 KB unified
physically addressed
virtual address space is 1 GB
fully associative
cache holds 20 KB
40 entries
5 way set associative
1 byte control/entry
32 Byte block size
sector size = block size

Advanced Computer Arc.

55

LRU replacement
write back
byte addresses on word boundaries
2a) Sketch a block diagram of how the virtual address is mapped into the physical
address (assuming a TLB hit). Be sure to label exactly which/how many of the
address bits go
where. and how many bits are in each of the 3 fields in a TLB entry.

1 GB MEANS
30-BIT
VIRTUAL
ADDRESS

4 KB PAGE SIZE MEANS LOW 12 BITS ARE

UNTRANSLATED
TLB
BITS 12-29

LOOKUP

18 bit
8 bits
14 bit
PAGE CONTROL
PAGE
NAME
ADDRESS

ASSOCIATIVE

14-BIT PAGE ADDRESS

(BITS 12-26 OF PHYSICAL ADDRESS)

Case Study 3. Multi-Level Caches.

You have a computer with two levels of cache memory and the following
specifications:
CPU Clock: 200 MHz Bus speed: 50 MHz
Processor: 32-bit RISC scalar CPU, single data address maximum per instruction
L1 cache on-chip, 1 CPU cycle access
block size = 32 bytes, 1 block/sector, split I & D cache
each single-ported with one block available for access, non-blocking
L2 cache off-chip, 3 CPU cycles transport time (L1 miss penalty)

56

block size = 32 bytes, 1 block/sector, unified single-ported cache, blocking, nonpipelined

Main memory has 12+4+4+4 CPU cycles transport time for 32 bytes (L2 miss
penalty)
Below are the results of a dinero simulation for the L1 cache:
CMDLINE: dinero -b32 -i8K -d8K -a1 -ww -An -W8 -B8
CACHE (bytes): blocksize=32, sub-blocksize=0, wordsize=8, Usize=0,
Dsize=8192, Isize=8192, bus-width=8.
POLICIES: assoc=1-way, replacement=l, fetch=d(1,0), write=w, allocate=n.
CTRL: debug=0, output=0, skipcount=0, maxcount=10000000, Q=0.
Metrics
Access Type:
(totals,fraction) Total
Instrn Data Read Write Misc
----------------- ------ ------ ----- ----- ----------Demand Fetches
10000000
7362210 2637790 1870945 766845
0
1.0000
0.7362
0.2638 0.1871 0.0767
0.0000
Demand Misses
52206 8466
43740 36764 6976 0
0.0052 0.0011 0.0166 0.0196 0.0091 0.0000
Words From Memory 180920
( / Demand Fetches) 0.0181
Words Copied-Back 766845
( / Demand Writes) 1.0000
Total Traffic (words) 947765
( / Demand Fetches) 0.0948
3a) What is the available (as opposed to used) sustained bandwidth between:
- L1 cache bandwidth available to CPU (assuming 0% L1 misses)?
200 MHz * 2 caches * 32 bytes / 1 clock = 12.8 * 109 B/sec = 11.92 GB/sec
- L2 cache bandwidth available to L1 cache (assuming 0% L2 misses)?
200 MHz * 1 cache * 32 bytes / 3 clocks = 2.133 * 109 B/sec = 1.98 GB/sec
- Main memory bandwidth available to L2 cache?
200 MHz * 32 bytes / (12+4+4+4) clocks = 267 * 106 B/sec = 254 MB/sec
3b) How long does an average instruction take to execute (in ns), assuming 1 clock
cycle per instruction in the absence of memory hierarchy stalls, no write buffering at
the L1 cache
level, and 0% L2 miss rate.
7362210 instructions = 7362210 clock cycles @ 1 clock effective access time
52206 demand misses @ 3 clocks = 156618 clocks delay penalty.
(7362210 + 156618) / 7362210 = 1.0213 clocks / 200 Mhz = 5.1065 ns
3c) A design study is performed to examine replacing the L2 cache with a victim
cache.

Advanced Computer Arc.

57

Compute a measure of speed for each alternative and indicate which is the faster
solution. Assume
the performance statistics are
L2 cache local miss ratio = 0.19
Victim cache miss ratio = 0.26; and its transport time from L1 miss = 1 clock
Given fixed L1 cache performance, it is fair to compare these head-to-head (but the
comparison
might not stay the same if L1 were changed):
tea for L2 cache beyond the L1 access time is:
3 + 0.19 * (12+4+4+4) = 7.56 clocks in addition to L1 delay
tea for L2 cache beyond the L1 access time is:
1 + 0.26 * (12+4+4+4) = 7.24 clocks in addition to L1 delay
So, in this (contrived) case the victim cache is a slight win in speed, and a whole lot
cheaper.

Case Study 4.
Discuss about the advantage(s) and the disadvantages of the von Nuemann concept.

Solution:
The von Neumann concept is a computer design model that uses a single storage
model to hold both instructions and data.
Advantages:
Reprogramming was made easier
Programs are allowed to modify themselves
Programs can write Programs
General flexibility
Disadvantages:
Malfunctioning programs can damage other programs or the operating
system
von Neumann bottleneck - CPU must wait for Data to transfer to and from
memory

58

Case Study 5.
Define term delayed branch, its application, and its shortcomings.

Solution:

Delayed branch is technique for reducing the effects of control dependencies by

delaying the point where a branch operation effects the program counter. This
allows one or more instructions following the branch operation to execute
whether or not the branch operation succeeds.

Advantage:
Allows for pipeline CPUs to reduce the clock cycles wasted due to pipeline
flushing during a branch or a jump operation

Disadvantage:
If the compiler cannot put instructions to execute after the branch due to
dependencies, then it must insert no-op instructions which increases the size
of program

Case Study 6.
CPU time (T) is defined as:
T = Ic* CPI *
Ic stands for the instruction count,

Advanced Computer Arc.

CPI stands for average clock cycles per instruction, and

stands for the clock cycle time.
A RISC computer, ideally, should be able to execute one instruction per clock
cycles. Within the scope of a RISC architecture, name and discuss (briefly)
distinct issues that do not allow ideal performance.

Solution:

Issues:
Memory Access: Any access to the memory can take longer than one
instruction
Branching: Program branches will flush instructions in a pipeline and cause it
to take longer then one instruction

Case Study 7. Loop fusion allows two or more loops that are executed the same
number of times and that use the same indices to be combined into one loop:

1. Within the scope of a RISC processor, why does it (Loop fusion) improve
performance (detail explanation)?

Solution:

In the scope of a RISC processor, Loop fusion can improve performance by

decreasing the need for extraneous loop control instructions. In the absence of
extraneous loop control instructions, the processor can run a program faster.

59

60

Case Study 8s. Interleave memory

1. Define interleaved memory (be as clear as possible);

Solution:

Interleaved memory describes a way to virtually access memory into a number of

memory banks.

2. Within the scope of interleaved memory, define mapping of the logical

addresses to the physical addresses. Distinguish them from each other.

Solution:
In interleaved memory, the memory is divided into N banks of memory where
virtual address, i, would actually reside in memory bank i/N (ignoring the
remainder), logically addressed by i mod N.

3. What is the main difference between an interleaved memory and a parallel

memory?

Solution:

Interleaved memory requires 2 to N memory banks to look up multiple contiguous

virtual memory locations where parallel memory only requires 1 memory
bank.

Advanced Computer Arc.

61

Multiple Choice Question

1.

2.

3.

Set-A
RISC stands for:
(a)
Register Instruction Set Computer
(b)
Reduced Instruction Set Computer
(c)
Reduced Instruction Set Clock
(d)
None of the above

( )

CISC processor have...............length instruction format.

(a)
Variable
(b)
Fixed
(c)
Can not say
(d)
None of the above

( )

CISC stands for:

(a)
Clock Instruction Set Computer
(b)
Control Instruction Set Computer
(c)
Complex Instruction Set Computer
(d)
None of the above

( )

4.

........................is a term used to denote a large class of techniques that are used to provide
simultaneous data processing tasks
(a)
Shared memory
(b)
Parallel Processing
(c)
Memory hierarchy
(d)
None of the above
( )

5.

...............................is a technique of decomposing a sequential process into sub operations.

(a)
Pipelining
(b)
Parallel Processing
(c)
Vector Processing
(d)
None of the above
( )

6.

Personal computer were appeared in:

(a)
Ist generation
(b)
2nd generation
(c)
4th generation
(d)
5th generation

7.

A.............contains the address of the next instructions to be executed.

(a)
Data Register
(b)
Accumulator

( )

62

(c)
(d)
8.

Instruction Register
Program Counter

( )

A......................is an interconnected set of processing elements which cooperate by

communicating with one another to solve large problem
(a)
Parallel Computer
(b)
Personal Computer
(c)
Laptop Computer
(d)
None of the above

( )

9.

Many operating system are designed to enable the CPU to process a number of
independent program concurrently. This concept is called:
(a)
Cache Memory
(b)
Multiprogramming
(c)
Multiprocessor
(d)
None of the above
( )

10.

A.......................system is an interconnection of two more CPU with memory and I/O

equipment.
(a)
Processor
(b)
Synchronization
(c)
Multiprocessor
(d)
None of the above
( )

11.

MIMD stands for:

(a)
More Instruction Stream, Multiple data Stream
(b)
Multiple Instruction Stream, Multiple Data Stream
(c)
Many Instruction Stream, Many Data Stream
(d)
None of the above

12.

The components that form a multiprocessor system are:

(a)
CPUs
(b)
IOPs
(c)
Memory Unit
(d)
All of the above

( )

( )

13.

Computers are interconnected with each other by means of communication lines to form
a:
(a)
Computer Network
(b)
Multiprocessor
(c)
Data Dependency
(d)
None of the above
( )

14.

A multiprocessor system with common shared memory is called:

(a)
Loosely coupled system
(b)
Tightly coupled system
(c)
Both a and b
(d)
None of the above

15.

Loosely coupled system are more efficient when the interaction between task is:

( )

Advanced Computer Arc.

(a)
(c)
16.

17.

Maximum
Can not say

63

(b)
(d)

Minimum
None of the above

( )

In a ......................has it own private local memory

(a)
Crossbar switch
(b)
Tightly coupled system
(c)
Loosely coupled system
(d)
None of the above

( )

The memory connected to the common system bus is.......................by all processors.
(a)
Shared
(b)
Partitioned
(c)
Distributed
(d)
None of the above

( )

18.

The ...........................organization consists of number of cross points that are placed at

intersection between buses and memory module paths
(a)
Multiport memory
(b)
Crossbar switch
(c)
Multistage switch
(d)
None of the above
( )

19.

A bus that connects components in a multiprocessor system, is called:

(a)
Control bus
(b)
Data bus
(c)
Address bus
(d)
System bus

( )

A typical system bus consists of approximately................signals lines

(a)
100
(b)
2
(c)
3
(d)
None of the above

( )

The signal lines in system bus are divided into.................functional groups.

(a)
1
(b)
2
(c)
3
(d)
None of the above

( )

The purpose of parallel processing is to:

(a)
Speed up the processing
(b)
Increase memory
(c)
Decrease memory
(d)
None of the above

( )

M.J. Flynn's parallel processing classification is based on:

(a)
Multiple Instructions
(b)
Multiple data
(c)
Both (a) and (b)
(d)
None of the above

( )

20.

21.

22.

23.

24.

..................................represents an organization that includes many processing units

under the supervision of a common control unit.
(a)
SISD
(b)
SIMD

64

(c)

MIMD

(d)

None of the above

( )

25.

A.....................can be be visualized as a collection of processing segments through which

binary information flows:
(a)
Memory
(b)
I/O devices
(c)
Processor
(d)
Pipeline
( )

26.

The pipeline used for floating point operations is called:

(a)
Arithmetic pipeline
(b)
Instruction pipeline
(c)
Both (a) and (b)
(d)
None of the above

27.

28.

29.

30.

( )

The finest level of pipelining is called:

(a)
Micro pipelining
(b)
Macro pipelining
(c)
Linear pipelining
(d)
None of the above

( )

VLIW stands for:

(a)
Very Long Instruction Word
(b)
Very Long Instruction Word
(c)
Very Large Information Word
(d)
None of the above

( )

MIPS stands for:

(a)
Memory Instruction Per Second
(b)
Major Instruction Per Second
(c)
Main Information Per Second
(d)
Million Instruction Per Second

( )

Router is a :
(a)
Data Transfer Protocol
(b)
Networking device
(c)
Modem
(d)
None of the above

( )

Answer
1. (b)
11. (b)
21. (c)

2. (b)
12. (d)
22. (a)

3. (c)
13. (a)
23. (c)

4. (b)
14. (b)
24. (b)

5. (a)
15. (b)
25. (a)

6. (c)
16. (c)
26. (a)

7. (d)
17. (a)
27. (a)

8. (a)
18. (a)
28. (a)

9. (b)
19. (d)
29. (d)

10. (c)
20. (a)
30. (b)

Advanced Computer Arc.

65

Set-B
1.

The channel width of a ..............network increases as we ascend from leaves to the root.
(a)
(b)
(c)
(d)

2.

3.

4.

5.

6.

7.

8.

Binary fat tree

Star
Ring
Binary tree

( )

The time required for two processes to synchronize with each other is called:
(a)
Synchronization time
(b)
Synchronization Latency
(c)
Process Latency
(d)
Memory latency

( )

If the number of links is 2N, then this would be which kind of network?
(a)
Illiac mesh
(b)
2D Mesh
(c)
Both (a) and (b)
(d)
None of the above

( )

TLB is used in:

(a)
Paging
(b)
Segmentation
(c)
Both (a) and (b)
(d)
None of the above

( )

The topology of an interconnection network can be:

(a)
Static
(b)
(c)
Either (a) and (b)
(d)

Dynamic
None of the above

( )

SIMD is a :
(a)
Memory management scheme
(b)
Processor for multiple organization
(c)
Attachment array processor
(d)
Programming technique

( )

The diameter of a network is the:

(a)
Maximum shortest path between any two nodes
(b)
Minimum shortest path between any two nodes
(c)
Minimum shortest path between any two adjacent nodes
(d)
Minimum longest path between any two adjacent nodes

( )

The size of program is determined of:

(a)
Clock Rate
(b)
Clock Count

66

(c)
(d)
9.

10.

11.

12.

13.

Instruction Execution Rate

Instruction count

( )

Multiprocessor is one with:

(a)
One CPU executing several processors
(b)
One CPU and several channels
(c)
Several CPU
(d)
None of the above
a crossbar switch network is a :
(a)
Regular connection network
(b)
Irregular connection network
(c)
Static connection network
(d)
Dynamic connection network

( )

( )

VLIW stands for:

(a)
Vector Large Instruction Word
(b)
Very Long Instruction Word
(c)
Very Large Integrated Word
(d)
Very Low Integrated Word

( )

The major disadvantage of pipeline is:

(a)
High cost individual dedicated
(b)
Initial setup time
(c)
If branch instruction is encountered the pipe has to be flushed
(d)
All of the above

( )

..................networks are controlled by a global clock

(a)
Asynchronous
(b)
Synchronous
(c)
Both (a) and (b)
(d)
Neither (a) nor (b)

( )

14.

.................is a shared memory system in which the access time varies with the location of
the memory word:
(a)
COMA
(b)
UMA
(c)
NUMA
(d)
All of the above
( )

15.

Which is not a valid data routing function?

(a)
Perfect shuffle and exchange
(b)
Permutation
(c)
Multicast
(d)
Broadband

16.

To find out cache performance, we can use:

( )

Advanced Computer Arc.

(a)
(b)
(c)
(d)
17.

18.

19.

20.

21.

22.

67

Program trace driven simulation

Hit Ratio
Creedy Cycles
Cycle count

( )

Which are the valid vector access memory schemes?

(a)
C-access Memory Organization
(b)
Synchronous Memory Organization
(c)
D-Access memory organization
(d)
Asynchronous memory organization
Which of the example of blocking network?
(a)
Baseline
(b)
Delta
(c)
Omega
(d)
All of the above

( )

( )

MAL stands for:

(a)
Minimal Average Latency
(b)
Minimum Allocation Latency
(c)
Maximum Allocation Latency
(d)
Maximum Average Latency

( )

The process of assigning control of the data transfer bus to a requester is called:
(a)
Interleaving
(b)
Interruption
(c)
Synchronization
(d)
Arbitration

( )

DOP stands for:

(a)
Dual Operating
(c)
Degree of processing

( )

(b)

Dual of Parallelism
(d)
Degree of parallelism

In general vector processing is faster and.................scalar processing:

(a)
Less efficient than
(b)
Equally efficient to
(c)
More efficient than
(d)
None of the above

( )

23.
Processors that use multiphase clock with a much increases clock rate ranging from 100 to
500 MHz.
(a)
RISC
(b)
VLIW
(c)
Both a and b
(d)
None of the above
( )

68

24.

..........................computing is achieved through the use of an array of processing elements

synchronized by the same controller;
(a)
MIMD
(b)
SIMD
(c)
Both a and b
(d)
None of the above
( )

25.

An/a .....................is a request from I/O or other devices to a processor for services or
attention:
(a)
Transaction
(b)
Arbitration
(c)
Interrupt
(d)
None of the above
( )

26.

The accumulated rate of all stage utilization determines:

(a)
Bounds of MAL
(b)
Pipeline throughout
(c)
Pipeline efficiency
(d)
Delay speed

( )

LRU stands for:

(a)
Last recently used
(b)
Least Recently used
(c)
Last Rarely Used
(d)
Least Rarely Used

( )

The memory hierarchy developed was based on a program behaviour is known is:
(a)
Locality of reference
(b)
Locality of coherence
(c)
Coherence property
(d)
None of the above

( )

27.

28.

29.

In a UMA multiprocessor model all processor have............access time to all memory

words:
(a)
Asynchronous
(b)
Equal
(c)
Different
(d)
None of the above
( )

30.

Throughout is measure of:

(a)
Number of instruction set executed per unit of time
(b)
Time for the completion of the task
(c)
Work done by the CPU
(d)
Memory Speed

( )

Answer
1. (a)
11. (b)
21. (d)

2. (b)
12. (d)
22. (c)

3. (c)
13. (b)
23. (c)

4. (a)
14. (c)
24. (b)

5. (c)
15. (d)
25. (c)

6. (b)
16. (b)
26. (b)

7. (a)
17. (a)
27. (b)

8. (d)
18. (d)
28. (a)

9. (a)
19. (a)
29. (b)

10. (a)
20. (b)
30. (c)

Advanced Computer Arc.

1.

69

Set C
Cache memory is:
(a)
Temporary and costly
(b)
Primary
(c)
High speed memory
(d)
All of the above

( )

2.

..................instruction is used to store the contents of accumulator into the memory word
specified by the effective addresses:
(a)
LDA
(b)
BUN
(c)
STA
(d)
BSA
( )

3.

What does RISC stand for?

(a)
Register Instruction Set Counter
(b)
Reduced Instruction Set Computer
(c)
Reduced Instruction Set Counter
(d)
Register Instruction Set Computer

( )

4.

A computer system consists of a CPU, a memory and one or more specialized I/O
processor called:
(a)
Bandwidth
(b)
Data Channels
(c)
Interrupt
(d)
None of the above
( )

5.

Which of the following is bus architecture:

(a)
ISA
(b)
AGA
(c)
MCA
(d)
All of the above

6.

7.

( )

SIMM is a :
(a)
Single Instruction memory modular
(b)
Single in Line Memory Modular
(c)
Single Instruction Memory Manufacturer
(d)
Single in Line Micrograms Modular

( )

Which of not an address mapping scheme:

(a)
Associate Mapping
(b)
Direct Mapping
(c)
Direct Associate Mapping
(d)
Set Associate Mapping

( )

70

8.

9.

10.

11.

12.

13.

14.

Example of zero address instruction is:

(a)
ADD B
(b)
ADD
(c)
ADD R1, B
(d)
ADD R1, A, B

( )

The speed of microcomputer measure in:

(a)
MIPS
(b)
Picoseconds
(c)
Megahertz
(d)
Milihertz

( )

The memory used in a computer system is based on the following principle:

(a)
Principle of parallel computing
(b)
Principle of concurrent occur
(c)
Principle of locality
(d)
None

( )

Instruction to be executed resides:

(a)
Program counter
(b)
Accumulator
(c)
MBR
(d)
Instruction Register

( )

One nibble is equivalent to:

(a)
8 bits
(b)
4 bytes
(c)
8 bytes
(d)
4 bits

( )

RAID is:
(a)
A computer
(b)
Storage device
(c)
A type of input device
(d)
A type of output device

( )

The instruction set of a.............processor usually contains 200 to 300 instructions:

(a)
CISC
(b)
SMP
(c)
RISC
(d)
All of the above

( )

Advanced Computer Arc.

15.

16.

17.

18.

19.

20.

21.

22.

71

The transformation of data from main memory to cache memory is called:

(a)
Mapping
(b)
Processing
(c)
Counting
(d)
Multiplexing

( )

A memory deices in which a bit is stored as a charge across the stray capacitance:
(a)
SRAM
(b)
EPROM
(c)
DRAM
(d)
Bubble Memory

( )

Multiprocessor is one with:

(a)
One CPU and several channels
(b)
Several CPU
(c)
One CPU executing several processors
(d)
None of the above

( )

The size of virtual memory depends on:

(a)
The size of data bus
(b)
The size of the main memory
(c)
The size of the address bus
(d)
None

( )

Swapping is also called:

(a)
Roll out technique
(b)
Roll in technique
(c)
Roll out roll in technique
(d)
All of the above

( )

..........is responsible for multiplexing the CPU:

(a)
Device manager
(b)
Device controller
(c)
Scheduler
(d)
None

( )

The term 'Baud Rate' is a measure of the :

(a)
Memory capacity
(b)
Speed at which data travels over the network
(c)
Instruction execution time
(d)
All of the above

( )

Two or more CPU's present in a computer system which share some or all of the memory
called:
(a)
Paralled
(b)
Multipgramming
(c)
Multi tasking
(d)
Random File processing ( )

72

23.

Which mode transits data in both directions, but not at the same time:
(a)
Simplex mode
(b)
Half duplex mode
(c)
Full duplex
(d)
None

( )

24.

If a process does not have direct and unassisted access to data items these items are
said to be:
(a)
Off line
(b)
Time shared
(c)
On line
(d)
None
( )

25.

Virtual memory is:

(a)
A part of main memory
(b)
Shared memory
(c)
Part of cache memory
(d)
A mechanism to process fast

( )

Throughout is a measures of:

(a)
Memory
(b)
Number of instruction sets executed per unit of time
(c)
Work done by the CPU
(d)
Time for the completion of the risk

( )

Pipelining is:
(a)
Programming technique
(b)
Decomposing of sequential process into sub operations
(c)
Hardware module
(d)
None

( )

Vector processing is not part of :

(a)
Multiprocessing
(b)
Parallel processing
(c)
Batch mode processing
(d)
Array processors

( )

26.

27.

28.

29.

30.

Memory interleaving is:

(a)
Modular memory
(b)
Virtual memory
(c)
Shared memory
(d)
Cache memory
SIMD is a :
(a)
Memory management scheme
(b)
Processor for multiple data organization
(c)
Attached array processor
(d)
Programming technique

( )

( )

Advanced Computer Arc.

73

Answer
1. (d)
11. (d)
21. (b)

2. (c)
12. (d)
22. (a)

3. (b)
13. (b)
23. (b)

4. (d)
14. (a)
24. (a)

5. (d)
15. (a)
25. (d)

6. (a)
16. (c)
26. (b)

7. (c)
17. (c)
27. (b)

8. (b)
18. (c)
28. (c)

9. (a)
19. (c)
29. (b)

10. (c)
20. (c)
30. (b)

74

Set D
1.

....................modules are used to select one out of n read or write requested for service:
(a)
Multiplexer
(b)
Demultiplexer
(c)
Crossbar
(d)
All of the above
( )

2.

In these types of computers, the execution on instruction is driven by data availability

insured of being guided by a program counter:
(a)
Control Flow Computer
(b)
Data Flow Computer
(c)
Both (a) and (b)
(d)
Neither (a) nor (b)
( )

3.

It manager the allocation and deal location of resources during the exeuction of user
programs:
(a)
CPU
(b)
OS
(c)
Monitor
(d)
Keyboard
( )

4.

Multi-processor, vector-supercomputer, and multi computer were appeared in:

(a)
1st generation
(b)
2nd generation
(c)
3rd generation
(d)
4th generation

( )

Full form of MIPS is:

(a)
Multi instructions
(b)
Million Instructions Per Second
(c)
Multi Instructions Processor System
(d)
None of the above

( )

The physical memory is uniformly shared by all the processors in:

(a)
NUMA
(b)
COMA
(c)
UMA
(d)
All of the above

( )

Which is not valid architecture of a vector super computer?

(a)
Register to register
(b)
Memory to memory
(c)
Both a and b are invalid
(d)
None of the above

( )

5.

6.

7.

Advanced Computer Arc.

75

8.

........................is a measure of the amount of compilation involved in a software process:

(a)
Grain size
(b)
Granularity
(c)
Both a and b
(d)
neither a nor b
( )

9.

The diameter of a network is the :

(a)
Maximum shortest path between any two nodes
(b)
Minimum shortest path between any two nodes
(c)
Maximum shortest path between any two adjacent nodes
(d)
Minimum longest path between two adjacent nodes

10.

11.

12.

( )

The channel width of a fat tree increase as we ascend form:

(a)
Leaves to the roots
(b)
Roots to the leaves
(c)
Width
(d)
None of the above

( )

The total number of messages the network can handle is:

(a)
Network efficiency
(b)
Network throughout
(c)
Network output
(d)
All of the above

( )

Which dynamic connection network has a low with limited bandwidth features?
(a)
Multistage interconnection
(b)
Crossbar switch
(c)
Bus system
(d)
All of the above

( )

13.

Which dynamic connection network provides highest bandwidth and interconnection?

(a)
Crossbar networks
(b)
Multi stage networks
(c)
Bus system
(d)
None of the above
( )

14.

DOP stands for:

(a)
Dual Operating Processor
(b)
Degree of parallelism
(c)
Dual of parallelism
(d)
Degree of processing

( )

15.
...........................is a cascade of processing stage which are linearly connected to perform a
fixed function over a stream of data flowing from one end to the other:
(a)
Omega network
(b)
Tree Network
(c)
Linear Pipeline
(d)
Parallel Processor `
( )
16.

The finest level of pipelining is called:

(a)
Macro pipelining

76

(b)
(c)
(d)
17.

Micro pipelining
Man pipelining
None of the above

( )

When the addressed data/instruction is found in cache it is called:

(a)
Cache hit
(b)
Cache miss
(c)
Cache found
(d)
Cache trace

( )

18.

In proposed future bus + standard, the..................are used to declare special bus

transaction:
(a)
Command Lines
(b)
Status lines
(c)
Capability lines
(d)
Miscellaneous lines
( )

19.

The percentage of time that each pipeline stage is used over a sufficiently long series of
task initiation is the:
(a)
Pipeline throughout
(b)
Pipeline efficiency
(c)
Pipeline latency
(d)
Pipeline clock
( )

20.

The effective bandwidth available to each processor is inversely proportional

to...........contending for the bus
(a)
Number of terminal
(b)
Number of processor
(c)
Number of operating systems
(d)
All of the above
( )

21.

Flynn's classification is based on:

(a)
Memory instructions
(b)
Hardware types
(c)
Software types
(d)
Notion of instruction and data streams

22.

23.

The size of program is determined by:

(a)
Clock Rate
(b)
Instruction Execution Rate
(c)
Instruction Count
(d)
Clock Count

( )

( )

The............model is a special case of NUMA machine, in which the distributed main

memories are converted to caches:
(a)
UMA
(b)
Special NUMA
(c)
COMA
(d)
None of the above
( )

Advanced Computer Arc.

77

24.

The................network provide point to point static connections among the nodes:

(a)
Message passing
(b)
Data Passing
(c)
Both a and b
(d)
Neither a nor b
( )

25.

MAL stands for:

(a)
Minimum allocation Latency
(b)
Minimal Average Latency
(c)
Maximum Average Latency
(d)
Maximum Allocation Latency

( )

Latencies that causes collision are called:

(a)
Minimum Allocation Latency
(b)
Minimal Average Latency
(c)
Maximum Average Latency
(d)
Maximum Allocation Latency

( )

26.

27.

.....................is a pair of nodes that accounts for a disproportionately large portion of the
total network traffic.
(a)
Hit ration
(b)
Hot ratio
(c)
Hot spot
(d)
Hit spot
( )

28.

The memory hierarchy developed was based on a program behaviour is known as:
(a)
Coherence property
(b)
Locality of reference
(c)
Locality of coherence
(d)
None of the above

( )

Write through and write back are two strategies for maintaining:
(a)
Locality of reference
(b)
Collision
(c)
Coherence
(d)
All of the above

( )

In paging, if demanded page is not found it is declared is:

(a)
Page miss
(b)
Page fault
(c)
Both a and b
(d)
Electronic data interchange for administration commerce and trade

( )

29.

30.

Answer
1. (a)
2. (b)
11. (b)
12. (c)
21. (d)
22. (c)

3. (b)
13. (a)
23. (b)

4. (d)
14. (b)
24. (a)

5. (b)
15. (c)
25. (b)

6. (c)
16. (b)
26. (c)

7. (d)
17. (a)
27. (c)

8. (a)
18. (a)
28. (b)

9. (a)
19. (c)
29. (d)

10. (a)
20. (b)
30. (b)

78

BACHELOR OF COMPUTER APPLICATIONS

(Part III) EXAMINATION
(Faculty of Science)
(Three Year Scheme of 10+2+3 Pattern)
PAPER 314

ADVANCED COMPUTER
ARCHITECTURE
OBJECTIVE PART- I

Year - 2011
Time allowed : One Hour

Maximum Marks : 20

The question paper contains 40 multiple choice questions with four choices and students
will have to pick the correct one. (Each carrying marks.).

1.

2.

3.

4.

TLB is used in:

(a)
Paging
(b)
Segmentation
(c)
Both (a) and (b)
(d)
None of the above

()

The size of Program is determined by:

(a)
Clock Rate
(b)
Clock Count
(c)
Instruction Execution Rate
(d)
Instruction count

( )

VLIW stands for:

(a)
Vector large instruction word
(b)
Very long instruction word
(c)
Very large integrated word
(d)
Very low integrated word

( )

..........................networks are controlled by a global clock.

(a)
Asynchronous
(b)
Synchronous
(c)
Both (a) and (b)

Advanced Computer Arc.

(d)
5.

6.

7.

8.

9.

10.

11.

12.

79

Neither (a) nor (b)

LRU stands for:

(a)
Last recently used
(c)
Last rarely used

( )

(b)
(d)

Least recently used

None of the above

( )

Router is a :
(a)
Data transfer protocol
(b)
Networking device
(c)
Modem
(d)
None of the above

( )

Personal computer were appeared in:

(a)
Ist generation
(b)
IInd generation
(c)
IVth generation
(d)
Vth generation

( )

Cache memory is:

(a)
Temporary and costly
(b)
Primary
(c)
High speed memory
(d)
All of the above

( )

SIMM is a :
(a)
Single Instruction Memory Modular
(b)
Single in-Line memory modular
(c)
Single instruction memory manufacturer
(d)
Single In-Line Micro programs Modular
The speed of microcomputer is measured in:
(a)
MIPS
(c)
Megahertz

(b)
(d)

( )

Picoseconds
Millihertz

( )

One nibble is equivalent to:

(a)
8-bits
(b)
4-bytes
(c)
8-bytes
(d)
4-bits
Virtual memory is:
(a)
A part of main memory
(b)
Shared memory
(c)
Part of cache memory
(d)
A mechanism to execute large size programs

( )

( )

80

13.

14.

15.

16.

17.

18.

19.

20.

Which gate is a universal gate?

(a)
NAND
(b)
OR
(c)
AND
(d)
None of the above
If a process does not have direct and unassisted access to data items, these items
said to be:
(a)
Off-line
(b)
Time shared
(c)
On-line
(d)
None of the above

( )
are

( )

Vector processing is not par to of:

(a)
Multiprocessing
(b)
Parallel processing
(c)
Batch mode processing
(d)
Array processors

( )

Swapping is also called:

(a)
Roll out technique
(b)
Roll in technique
(c)
Roll out roll in technique
(d)
All of the above

( )

Full form of MIPS is:

(a)
Multi Instructions Per Second
(b)
Million Instructions Per Second
(c)
Million Instruction Processor System
(d)
None of the above

( )

The total number of message the network can handle in:

(a)
Network Efficiency
(b)
Network throughput
(c)
Network output
(d)
All of the above

( )

Flynn's classification is based on:

(a)
Memory Instruction
(b)
Hardware types
(c)
Software types
(d)
Notation of instructions and data streams

( )

In Paging, if demanded page is not found it is declared as:

(a)
Page miss
(b)
Page Fault

Advanced Computer Arc.

(c)
(d)
21.

22.

23.

24.

25.

26.

27.

Both (a) and (b)

Neither (a) nor (b)

81

( )

Omega Network is a....................network.

(a)
High stage
(b)
Low stage
(c)
Single stage
(d)
Multistage

( )

For a reservation table with n columns, the maximum forbidden latency (m) :
(a)
m n
(b)
m nn
(c)
m n
(d)
m n1

( )

Switching complexity of multistage network is:

(a)
O (logk n)
(b)
O( )
(c)
O (n logk n)
(d)
O (n)

( )

The full form of PRAM is:

(a)
Parallel Random Access Machine
(b)
Parallel Remote Access Machine
(c)
Powerful Random Access Memory
(d)
Partial Random Access Memory

( )

In Binary tree network the bisection width would be:

(a)
1
(b)
2
(c)
N/2
(d)
(N/2) 2

( )

The T3D is a :
(a)
SIMD machine
(b)
MIMD machine
(c)
Both (a) and (b)
(d)
None of the above

( )

In future Bus + standard, the 64-bit address line are multiplexed with:
(a)
Lower order 64-bit data lines
(b)
High order 64-bit data lines
(c)
Lower order 32-bit data lines
(d)
Any one of the above

( )

82

28.

29.

30.

31.

32.

The dimensions of locality property are:

(a)
Temporal, parallel and sequential
(b)
Temporal, spatial and sequential
(c)
Spatial, parallel and sequential
(d)
None of the above

( )

Pipelining is:
(a)
Programming Techniques
(b)
Decomposing of sequential process into sub operations
(c)
Hardware module
(d)
None of the above

( )

Which is example of blocking network?

(a)
Baseline
(b)
Omega
(c)
Delta
(d)
All of the above

( )

Bootstrap is:
(a)
A memory device
(c)
Error correction technique

(b)
(d)

A startup program
None of the above

Only....................of the instruction of a complex instruction set are frequently

about 95% of the time:
(a)
75%
(b)
25%
(c)
80%
(d)
30%

( )
used

( )

33.

An asynchronous bus allow.......................device(s) to be connected on the same

bus.
(a)
Fast
(b)
Slow
(c)
Fast and Slow
(d)
None of the above
( )

34.

Ideally, linear pipeline of k stage can process n tasks in.......................Time:

(a)
[n + ( k 1 )]
(b)
[n + (k + 1)]
(c)
[k + (n 1)]
(d)
[k (n + 1)]

( )

Which of the following is/are representative CISC scalar processor?

(a)
Intel i 486
(b)
Motorola MC 68040
(c)
NS 32532
(d)
All of the above

( )

35.

Advanced Computer Arc.

36.

37.
be

38.

39.

40.

83

Which of the following is/are supercomputer?

(a)
Cray Y-MP
(b)
NEC SX2
(c)
Fujitsu VP 400
(d)
All of the above

( )

In Pipelined instruction Processing the.................process the instruction function

performed and identified the resources needed.
(a)
Fetch stage
(b)
Decode stage
(c)
Issue stage
(d)
Write back stage
A crossbar switch network is a/an:
(a)
Dynamic Connection Network
(b)
Static Connection Network
(c)
Regular Connection Network
(d)
Irregular Connection Network

to

( )

( )

The problem of aliasing is associated with...............cache.

(a)
Physical address
(b)
Virtual address
(c)
Split cache
(d)
Both (a) and (b)

( )

CPI stands for:

(a)
Cycle Per Instruction
(b)
Communication Process Interconnect
(c)
Calls Per instruction
(d)
None of the above

( )

______________

84

DESCRIPTIVE PART-II
Year- 2011
Time allowed : 2 Hours

Maximum Marks : 30

Attempt any four descriptive types of questions out of the six. All questions carry 7 marks
each.
1.

Define the following:

(i)
BUS;
(ii)
Primary and Secondary Memory;
(iii)
RAM and ROM;
(iv)
Processor;
(v)
Cache Memory

2.

(a)
(b)

What do you mean by pipelining? How do improve performance of a

computer system using pipelining?
Compare Linear and non-linear pipelined processor.

3.

(a)
(b)

Compare CISC and RISC.

Write a short note on Shared Memory?

4.

(a)
(b)

What are various conditions of Parallelism.

Write short note on Static Interconnection Network and Dynamic
Interconnection Network.

5.

(a)
(b)

Explain virtual memory management technology?

Write a short note on Vector on Processing.

6.

Write short notes on the following:

(a)
SIMD computers;
(b)
Program Flow mechanisms;
(c)
Vector VLIW and symbolic processors.

Advanced Computer Arc.

85

BACHELOR OF COMPUTER APPLICATIONS

(Part III) EXAMINATION
(Faculty of Science)
(Three Year Scheme of 10+2+3 Pattern)
PAPER 314

ADVANCED COMPUTER
ARCHITECTURE
OBJECTIVE PART- I

Year - 2010
Time allowed : One Hour

Maximum Marks : 20

The question paper contains 40 multiple choice questions with four choices and students
will have to pick the correct one. (Each carrying marks.).
1.

2.

3.

4.

RISC stands for:

(a)
Register Instruction Set Computer
(b)
Reduced Instruction Set Computer
(c)
Reduced Instruction Set Clock
(d)
None of the above

( )

CISC processor have...............length instruction format.

(a)
Variable
(b)
Fixed
(c)
Can not say
(d)
None of the above

( )

CISC stands for:

(a)
Clock Instruction Set Computer
(b)
Control Instruction Set Computer
(c)
Complex Instruction Set Computer
(d)
None of the above

( )

........................is a term used to denote a large class of techniques that are used to provide
simultaneous data processing tasks
(a)
Shared memory
(b)
Parallel Processing
(c)
Memory hierarchy
(d)
None of the above
( )

86

5.

...............................is a technique of decomposing a sequential process into sub operations.

(a)
Pipelining
(b)
Parallel Processing
(c)
Vector Processing
(d)
None of the above
( )

6.

Personal computer were appeared in:

(a)
Ist generation
(b)
2nd generation
(c)
4th generation
(d)
5th generation

( )

A.............contains the address of the next instructions to be executed.

(a)
Data Register
(b)
Accumulator
(c)
Instruction Register
(d)
Program Counter

( )

A......................is an interconnected set of processing elements which cooperate by

communicating with one another to solve large problem
(a)
Parallel Computer
(b)
Personal Computer
(c)
Laptop Computer
(d)
None of the above

( )

7.

8.

9.

Many operating system are designed to enable the CPU to process a number of
independent program concurrently. This concept is called:
(a)
Cache Memory
(b)
Multiprogramming
(c)
Multiprocessor
(d)
None of the above
( )

10.

A.......................system is an interconnection of two more CPU with memory and I/O

equipment.
(a)
Processor
(b)
Synchronization
(c)
Multiprocessor
(d)
None of the above
( )

11.

MIMD stands for:

(a)
More Instruction Stream, Multiple data Stream
(b)
Multiple Instruction Stream, Multiple Data Stream
(c)
Many Instruction Stream, Many Data Stream
(d)
None of the above

12.

The components that form a multiprocessor system are:

(a)
CPUs
(b)
IOPs
(c)
Memory Unit
(d)
All of the above

( )

( )

Advanced Computer Arc.

87

13.

Computers are interconnected with each other by means of communication lines to form
a:
(a)
Computer Network
(b)
Multiprocessor
(c)
Data Dependency
(d)
None of the above
( )

14.

A multiprocessor system with common shared memory is called:

(a)
Loosely coupled system
(b)
Tightly coupled system
(c)
Both a and b
(d)
None of the above

( )

Loosely coupled system are more efficient when the interaction between task is:
(a)
Maximum
(b)
Minimum
(c)
Can not say
(d)
None of the above

( )

In a ......................has it own private local memory

(a)
Crossbar switch
(b)
Tightly coupled system
(c)
Loosely coupled system
(d)
None of the above

( )

The memory connected to the common system bus is.......................by all processors.
(a)
Shared
(b)
Partitioned
(c)
Distributed
(d)
None of the above

( )

15.

16.

17.

18.

The ...........................organization consists of number of cross points that are placed at

intersection between buses and memory module paths
(a)
Multiport memory
(b)
Crossbar switch
(c)
Multistage switch
(d)
None of the above
( )

19.

A bus that connects components in a multiprocessor system, is called:

(a)
Control bus
(b)
Data bus
(c)
Address bus
(d)
System bus

( )

A typical system bus consists of approximately................signals lines

(a)
100
(b)
2
(c)
3
(d)
None of the above

( )

The signal lines in system bus are divided into.................functional groups.

(a)
1
(b)
2
(c)
3
(d)
None of the above

( )

20.

21.

22.

The purpose of parallel processing is to:

88

(a)
(b)
(c)
(d)
23.

Speed up the processing

Increase memory
Decrease memory
None of the above

( )

M.J. Flynn's parallel processing classification is based on:

(a)
Multiple Instructions
(b)
Multiple data
(c)
Both (a) and (b)
(d)
None of the above

( )

24.

..................................represents an organization that includes many processing units

under the supervision of a common control unit.
(a)
SISD
(b)
SIMD
(c)
MIMD
(d)
None of the above
( )

25.

A.....................can be be visualized as a collection of processing segments through which

binary information flows:
(a)
Memory
(b)
I/O devices
(c)
Processor
(d)
Pipeline
( )

26.

The pipeline used for floating point operations is called:

(a)
Arithmetic pipeline
(b)
Instruction pipeline
(c)
Both (a) and (b)
(d)
None of the above

27.

28.

29.

30.

( )

The finest level of pipelining is called:

(a)
Micro pipelining
(b)
Macro pipelining
(c)
Linear pipelining
(d)
None of the above

( )

VLIW stands for:

(a)
Very Long Instruction Word
(b)
Very Long Instruction Word
(c)
Very Large Information Word
(d)
None of the above

( )

MIPS stands for:

(a)
Memory Instruction Per Second
(b)
Major Instruction Per Second
(c)
Main Information Per Second
(d)
Million Instruction Per Second
Router is a :
(a)
Data Transfer Protocol
(b)
Networking device

( )

Advanced Computer Arc.

(c)
(d)

89

Modem
None of the above

( )

31.

Two or more CPUs present in a computer system which share some or all of the memory?
(a)
Parallel Processing
(b)
Multiprogramming
(c)
Random file processing
(d)
Multitasking
( )

32.

Flynn's classified parallel computers into..............categories.

(a)
2
(b)
3
(c)
4
(d)
8

( )

Which of the following is an interconnection network?

(a)
Time shared common bus
(b)
Crossbar switch
(c)
Multistage switch
(d)
All of the above

( )

Through is measure of:

(a)
Number of instruction set executed per unit of time
(b)
Time for the completion of the task
(c)
Work done by the CPU
(d)
Memory speed

( )

DOP stands for:

(a)
Dual Operating Processor
(b)
Dual of Parallelism
(c)
Degree of Processing
(d)
Memory speed

( )

..........................Networking are controlled by a global clock.

(a)
Asynchronous
(b)
Synchronous
(c)
Both (a) and (b)
(d)
None of the above

( )

Multiprocessor is one with:

(a)
One CPU executing several processor
(b)
Several CPU
(c)
One CPU and several channels
(d)
None of the above

( )

33.

34.

35.

36.

37.

38.

The topology of an interconnection network can be:

(a)
Dynamic
(b)
(c)
Both a and b
(d)

Static
None of the above

( )

90

39.

40.

To find out cache performance we can use:

(a)
Program trace driven simulation
(b)
Hit ration
(c)
Greedy cycles
(d)
Cycle Count

( )

..........................is a concept used in some large computer systems that permit the user to
construct programs as thought a large memory space.
(a)
Cache memory
(b)
Random access memory
(c)
Virtual memory
(d)
None of the above
( )

Answer Key

1. (b)

2. (b)

3. (c)

4. (b)

5. (a)

6. (c)

7. (d)

8. (a)

9. (b)

10. (c)

11. (b)

12. (d)

13. (a)

14. (b)

15. (b)

16. (c)

17. (a)

18. (a)

19. (d)

20. (a)

21. (c)

22. (a)

23. (c)

24. (b)

25. (a)

26. (a)

27. (a)

28. (a)

29. (d)

30. (b)

31. (d)

32. (c)

33. (b)

34. (a)

35. (d)

36. (b)

37. (b)

38. (c)

39. (b)

40. (c)

___________

Advanced Computer Arc.

91

DESCRIPTIVE PART-II

Year- 2010
Time allowed : 2 Hours

Maximum Marks : 30

Attempt any four descriptive types of questions out of the six. All questions carry 7 marks
each.

Q.1

(a)
(b)

What is meant by Parallelism? What are conditions of parallelism? Discuss

the
network properties.
Explain the static interconnection network.

Q.2

Write short notes on:

(a)
Superscalar
(b)
Memory hierarchy technology

Q.3

Differentiate the following :

(a)
Linear and non- linear Pipeline Processors;
(b)
Static and Dynamic Interconnection network.

Q.4

(a)
(b)

What is meant by cache memory? Discuss the virtual memory technology

Write short note on shared memory

Q.5

(a)
(b)

What is meant by pipelining? Discuss the instruction pipeline design.

Explain RISC processors.

Q.6

Write
(a)
(b)
(c)

short notes on:

Buses;
Multi vector Multiprocessors;
VLIW
________

92

OBJECTIVE PART- I
Year - 2009
Time allowed : One Hour

Maximum Marks : 20

The question paper contains 40 multiple choice questions with four choices and student will have
to pick the correct one. (Each carrying marks.).
1.

The channel width of a ..............network increases as we ascend from leaves to the root.
(a)
(b)
(c)
(d)

2.

3.

4.

5.

6.

Binary fat tree

Star
Ring
Binary tree

( )

The time required for two processes to synchronize with each other is called:
(a)
Synchronization time
(b)
Synchronization Latency
(c)
Process Latency
(d)
Memory latency

( )

If the number of links is 2N, then this would be which kind of network?
(a)
Illiac mesh
(b)
2D Mesh
(c)
Both (a) and (b)
(d)
None of the above

( )

TLB is used in:

(a)
Paging
(b)
Segmentation
(c)
Both (a) and (b)
(d)
None of the above

( )

The topology of an interconnection network can be:

(a)
Static
(b)
(c)
Either (a) and (b)
(d)
SIMD is a :
(a)
Memory management scheme
(b)
Processor for multiple organization
(c)
Attachment array processor
(d)
Programming technique

Dynamic
None of the above

( )

( )

Advanced Computer Arc.

7.

8.

9.

10.

11.

12.

13.

14.

93

The diameter of a network is the:

(a)
Maximum shortest path between any two nodes
(b)
Minimum shortest path between any two nodes
(c)
Minimum shortest path between any two adjacent nodes
(d)
Minimum longest path between any two adjacent nodes

( )

The size of program is determined of:

(a)
Clock Rate
(b)
Clock Count
(c)
Instruction Execution Rate
(d)
Instruction count

( )

Multiprocessor is one with:

(a)
One CPU executing several processors
(b)
One CPU and several channels
(c)
Several CPU
(d)
None of the above

( )

a crossbar switch network is a :

(a)
Regular connection network
(b)
Irregular connection network
(c)
Static connection network
(d)
Dynamic connection network

( )

VLIW stands for:

(a)
Vector Large Instruction Word
(b)
Very Long Instruction Word
(c)
Very Large Integrated Word
(d)
Very Low Integrated Word

( )

The major disadvantage of pipeline is:

(a)
High cost individual dedicated
(b)
Initial setup time
(c)
If branch instruction is encountered the pipe has to be flushed
(d)
All of the above

( )

..................networks are controlled by a global clock

(a)
Asynchronous
(b)
Synchronous
(c)
Both (a) and (b)
(d)
Neither (a) nor (b)

( )

.................is a shared memory system in which the access time varies with the location of
the memory word:
(a)
COMA
(b)
UMA

94

(c)
(d)
15.

16.

17.

18.

19.

20.

21.

22.

NUMA
All of the above

( )

Which is not a valid data routing function?

(a)
Perfect shuffle and exchange
(b)
Permutation
(c)
Multicast
(d)
Broadband

( )

To find out cache performance, we can use:

(a)
Program trace driven simulation
(b)
Hit Ratio
(c)
Creedy Cycles
(d)
Cycle count

( )

Which are the valid vector access memory schemes?

(a)
C-access Memory Organization
(b)
Synchronous Memory Organization
(c)
D-Access memory organization
(d)
Asynchronous memory organization

( )

Which of the example of blocking network?

(a)
Baseline
(b)
Delta
(c)
Omega
(d)
All of the above

( )

MAL stands for:

(a)
Minimal Average Latency
(b)
Minimum Allocation Latency
(c)
Maximum Allocation Latency
(d)
Maximum Average Latency

( )

The process of assigning control of the data transfer bus to a requester is called:
(a)
Interleaving
(b)
Interruption
(c)
Synchronization
(d)
Arbitration

( )

DOP stands for:

(a)
Dual Operating
(c)
Degree of processing

( )

(b)

Dual of Parallelism
(d)
Degree of parallelism

In general vector processing is faster and.................scalar processing:

(a)
Less efficient than
(b)
Equally efficient to

Advanced Computer Arc.

(c)
(d)

More efficient than

None of the above

95

( )

23.
Processors that use multiphase clock with a much increases clock rate ranging from 100 to
500 MHz.
(a)
RISC
(b)
VLIW
(c)
Both a and b
(d)
None of the above
( )
24.

..........................computing is achieved through the use of an array of processing elements

synchronized by the same controller;
(a)
MIMD
(b)
SIMD
(c)
Both a and b
(d)
None of the above
( )

25.

An/a .....................is a request from I/O or other devices to a processor for services or
attention:
(a)
Transaction
(b)
Arbitration
(c)
Interrupt
(d)
None of the above
( )

26.

The accumulated rate of all stage utilization determines:

(a)
Bounds of MAL
(b)
Pipeline throughout
(c)
Pipeline efficiency
(d)
Delay speed

( )

LRU stands for:

(a)
Last recently used
(b)
Least Recently used
(c)
Last Rarely Used
(d)
Least Rarely Used

( )

The memory hierarchy developed was based on a program behaviour is known is:
(a)
Locality of reference
(b)
Locality of coherence
(c)
Coherence property
(d)
None of the above

( )

27.

28.

29.

In a UMA multiprocessor model all processor have............access time to all memory

words:
(a)
Asynchronous
(b)
Equal
(c)
Different
(d)
None of the above
( )

30.

Throughout is measure of:

(a)
Number of instruction set executed per unit of time

96

(b)
(c)
(d)
31.

32.

Time for the completion of the task

Work done by the CPU
Memory Speed

( )

A asynchronous bus allows..............devices to be connected on the same bus:

(a)
Fast
(b)
Fast and flow
(c)
Broadway
(d)
Slow

( )

Omega network is a .................network:

(a)
High Stage
(b)
Low Stage
(c)
Single Stage
(d)
Multistage

( )

33.

......................is a pair of nodes that accounts for a disproportionately large portion of the
total network traffic.
(a)
Hot spot
(b)
Hit Spot
(c)
Hot ratio
(d)
Hit ratio
( )

34.

MIPS stands for:

(a)
Memory Instructions Per Second
(b)
Major Instructions Per Second
(c)
Main Instructions Per Second
(d)
Million Instructions Per Second

35.

36.

The finest level of pipelining is called:

(a)
Micro pipelining
(c)
Linear Pipeline processor
Router is a :
(a)
Data Transfer protocol
(b)
Networking device
(c)
Modem
(d)
None of the above

( )

(b)
(d)

Macro pipelining
None of the above

( )

( )

37.

Two or more CPUs present in a computer system which share some or all of the memory:
(a)
Parallel Processing
(b)
Multiprogramming
(c)
Random file processing
(d)
Multitasking
( )

38.

A non-linear pipeline allows:

Advanced Computer Arc.

(a)
(b)
(c)
(d)
39.

40.

97

Vector large instruction word

Very low integrated word
Very large integrated word
Very long instruction word

( )

Which of not an address mapping scheme:

(a)
Associate mapping
(b)
Direct Associate Mapping
(c)
Direct Mapping
(d)
Set Associate Mapping
Personal computers were appeared in:
(a)
Ist generation
(b)
2nd generation
(c)
4th generation
(d)
5th generation

( )

( )

Answer Key

1. (a)

2. (b)

3. (c)

4. (a)

5. (c)

6. (b)

7. (a)

8. (d)

9. (a)

10. (a)

11. (b)

12. (d)

13. (b)

14. (c)

15. (d)

16. (b)

17. (a)

18. (d)

19. (a)

20. (b)

21. (d)

22. (c)

23. (c)

24. (b)

25. (c)

26. (b)

27. (b)

28. (a)

29. (b)

30. (c)

31. (b)

32. (d)

33. (b)

34. (d)

35. (b)

36. (b)

37. (b)

38. (d)

39. (c)

40. (c)

_____________

98

DESCRIPTIVE PART - II
Year 2009
Time allowed : 2 Hours

Maximum Marks : 30

Attempt any four questions out of the six. All questions carry 7 marks each.

Q.1

(a)
(b)

Discuss and describe a typical superscalar architecture for a RISC processor.

What do you understand by superscalar and vector processes?

Q.2

(a)
(b)

Distinguish between multiprocessors and multicomputer.

Describe the UMA; NUMA and COMA shared memory multiprocessor models.

Q.3

Discuss the following cache performance issues:

(a)
Hit Ratio
(b)
Effect of block size
(c)
Cycle Counts.

Q.4

(a)
(b)

Discuss the terms data transfer bus (DTB) bus arbitration and control and
Financial modules related to backplane bus.
What do you mean by Backplane Bus system?

Q.5

(a)
(b)

What do you mean by dynamic interconnection network?

Discuss various factors which affect the performance of an interconnection
network.

Q.6

What is the basic difference between a linear pipeline processor and non-linear pipeline
processor? Discuss the asynchronous and synchronous models of linear pipeline
processor.
_______

Advanced Computer Arc.

99

OBJECTIVE PART- I
Year - 2008
Time allowed : One Hour

Maximum Marks : 20

The question paper contains 40 multiple choice questions with four choices and student will have
to pick the correct one. (Each carrying marks.).
1.

Cache memory is:

(a)
Temporary and costly
(b)
Primary
(c)
High speed memory
(d)
All of the above

( )

2.

..................instruction is used to store the contents of accumulator into the memory word
specified by the effective addresses:
(a)
LDA
(b)
BUN
(c)
STA
(d)
BSA
( )

3.

What does RISC stand for?

(a)
Register Instruction Set Counter
(b)
Reduced Instruction Set Computer
(c)
Reduced Instruction Set Counter
(d)
Register Instruction Set Computer

( )

4.

A computer system consists of a CPU, a memory and one or more specialized I/O
processor called:
(a)
Bandwidth
(b)
Data Channels
(c)
Interrupt
(d)
None of the above
( )

5.

Which of the following is bus architecture:

(a)
ISA
(b)
AGA
(c)
MCA
(d)
All of the above

6.

SIMM is a :
(a)
Single Instruction memory modular
(b)
Single in Line Memory Modular

( )

100

(c)
(d)
7.

8.

9.

10.

11.

12.

13.

14.

Single Instruction Memory Manufacturer

Single in Line Micrograms Modular

( )

Which of not an address mapping scheme:

(a)
Associate Mapping
(b)
Direct Mapping
(c)
Direct Associate Mapping
(d)
Set Associate Mapping

( )

Example of zero address instruction is:

(a)
ADD B
(b)
ADD
(c)
ADD R1, B
(d)
ADD R1, A, B

( )

The speed of microcomputer measure in:

(a)
MIPS
(b)
Picoseconds
(c)
Megahertz
(d)
Milihertz

( )

The memory used in a computer system is based on the following principle:

(a)
Principle of parallel computing
(b)
Principle of concurrent occur
(c)
Principle of locality
(d)
None

( )

Instruction to be executed resides:

(a)
Program counter
(b)
Accumulator
(c)
MBR
(d)
Instruction Register

( )

One nibble is equivalent to:

(a)
8 bits
(b)
4 bytes
(c)
8 bytes
(d)
4 bits

( )

RAID is:
(a)
A computer
(b)
Storage device
(c)
A type of input device
(d)
A type of output device

( )

The instruction set of a.............processor usually contains 200 to 300 instructions:

Advanced Computer Arc.

(a)
(b)
(c)
(d)
15.

16.

17.

18.

19.

20.

21.

CISC
SMP
RISC
All of the above

101

( )

The transformation of data from main memory to cache memory is called:

(a)
Mapping
(b)
Processing
(c)
Counting
(d)
Multiplexing

( )

A memory deices in which a bit is stored as a charge across the stray capacitance:
(a)
SRAM
(b)
EPROM
(c)
DRAM
(d)
Bubble Memory

( )

Multiprocessor is one with:

(a)
One CPU and several channels
(b)
Several CPU
(c)
One CPU executing several processors
(d)
None of the above

( )

The size of virtual memory depends on:

(a)
The size of data bus
(b)
The size of the main memory
(c)
The size of the address bus
(d)
None

( )

Swapping is also called:

(a)
Roll out technique
(b)
Roll in technique
(c)
Roll out roll in technique
(d)
All of the above

( )

..........is responsible for multiplexing the CPU:

(a)
Device manager
(b)
Device controller
(c)
Scheduler
(d)
None

( )

The term 'Baud Rate' is a measure of the :

(a)
Memory capacity
(b)
Speed at which data travels over the network
(c)
Instruction execution time
(d)
All of the above

( )

102

22.

Two or more CPU's present in a computer system which share some or all of the memory
called:
(a)
Paralled
(b)
Multipgramming
(c)
Multi tasking
(d)
Random File processing
( )

23.

Which mode transits data in both directions, but not at the same time:
(a)
Simplex mode
(b)
Half duplex mode
(c)
Full duplex
(d)
None

( )

24.

If a process does not have direct and unassisted access to data items these items are
said to be:
(a)
Off line
(b)
Time shared
(c)
On line
(d)
None
( )

25.

Virtual memory is:

(a)
A part of main memory
(b)
Shared memory
(c)
Part of cache memory
(d)
A mechanism to process fast

( )

Throughout is a measures of:

(a)
Memory
(b)
Number of instruction sets executed per unit of time
(c)
Work done by the CPU
(d)
Time for the completion of the risk

( )

Pipelining is:
(a)
Programming technique
(b)
Decomposing of sequential process into sub operations
(c)
Hardware module
(d)
None

( )

Vector processing is not part of :

(a)
Multiprocessing
(b)
Parallel processing
(c)
Batch mode processing
(d)
Array processors

( )

Memory interleaving is:

(a)
Modular memory
(b)
Virtual memory
(c)
Shared memory
(d)
Cache memory

( )

26.

27.

28.

29.

Advanced Computer Arc.

30.

31.

32.

33.

34.

35.

36.

37.

103

SIMD is a :
(a)
Memory management scheme
(b)
Processor for multiple data organization
(c)
Attached array processor
(d)
Programming technique

( )

Multi-programming is:
(a)
A technique to perform more than one task in memory
(b)
Capability to keep more than one program
(c)
A technique to perform paralel processing
(d)
None of the above

( )

The major disadvantages of pipelining is :

(a)
Initial setup time
(b)
If branch instruction is encountered the pipe has to be flushed
(c)
High cost of individual dedicated
(d)
All of the above

( )

Bootstrap is:
(a)
A memory device
(b)
a device to support the computer
(c)
A startup correction technique
(d)
an error correction technique

( )

Asynchronous communication protocols:

(a)
Transmit data in blocks
(b)
Serial data transfer
(c)
Continuous data transfer
(d)
Discontinuous data transfer

( )

Router is a:
(a)
Networking device
(b)
Data transfer protocol
(c)
Modem
(d)
None

( )

Bus is a:
(a)
Logical Channel for transferring data
(b)
Electronic pathway for transferring data
(c)
Dedicated path
(d)
None

( )

Which one not deal in network component ?

(a)
Novel
(b)
CISCO
(c)
Dell

104

(d)
38.

39.

40.

Microsoft

( )

Which gate is a universal gate?

(a)
NAND
(b)
OR
(c)
AND
(d)
None

( )

Which company does no deal with computer hardware?

(a)
IBM
(b)
Microsoft
(c)
HCL
(d)
Zenith

( )

The size of program is determined by:

(a)
Clock rate
(c)
Instruction execution

(b)
(d)

Clock count
Instruction count

( )

Answer Key
1. (d)
11. (d)
21. (b)
31. (a)

2. (c)
12. (d)
22. (a)
32. (d)

3. (b)
13. (b)
23. (b)
33. (c)

4. (d)
14. (a)
24. (a)
34. (a)

5. (d)
15. (a)
25. (d)
35. (a)

6. (a)
16. (c)
26. (b)
36. (a)

_____________

7. (c)
17. (c)
27. (b)
37. (c)

8. (b)
18. (c)
28. (c)
38. (a)

9. (a)
19. (c)
29. (b)
39. (b)

10. (c)
20. (c)
30. (b)
40. (d)

Advanced Computer Arc.

105

DESCRIPTIVE PART - II
Year 2008
Time allowed : 2 Hours
: 30

Maximum Marks

Attempt any four questions out of the six. All questions carry 7 marks each.

Q.1

What do you understand with virtual memory? Describe the page replacement techniques
in virtual memory

Q.2

Write short notes on the following:

(a)
Memory Hierarchy
(b)
Characteristics of multiprocessors

Q.3

Explain the interconnection structures of network. Describe the multistage switching

network with the help of diagram.

Q.4

What do you understand with pipelining processor? Describe the arithmetic pipeli ne with
the help of example

Q.5

Describe:
(i)
Vector Processing principles
(ii)
Memory interleaving

Q.6

Differentiate between the RISC and CISC processors. Explain also the characteristics of
both.

_______

106

OBJECTIVE PART- I
Year - 2007
Time allowed : One Hour

Maximum Marks : 20

The question paper contains 40 multiple choice questions with four choices and student will have
to pick the correct one. (Each carrying marks.).
1.

....................modules are used to select one out of n read or write requested for service:
(a)
Multiplexer
(b)
Demultiplexer
(c)
Crossbar
(d)
All of the above
( )

2.

In these types of computers, the execution on instruction is driven by data availabili ty

insured of being guided by a program counter:
(a)
Control Flow Computer
(b)
Data Flow Computer
(c)
Both (a) and (b)
(d)
Neither (a) nor (b)
( )

3.

It manager the allocation and deal location of resources during the exeuction of user
programs:
(a)
CPU
(b)
OS
(c)
Monitor
(d)
Keyboard
( )

4.

Multi-processor, vector-supercomputer, and multi computer were appeared in:

(a)
1st generation
(b)
2nd generation
(c)
3rd generation
(d)
4th generation

( )

Full form of MIPS is:

(a)
Multi instructions
(b)
Million Instructions Per Second
(c)
Multi Instructions Processor System
(d)
None of the above

( )

5.

6.

The physical memory is uniformly shared by all the processors in:

(a)
NUMA

Advanced Computer Arc.

(b)
(c)
(d)
7.

COMA
UMA
All of the above

Which is not valid architecture of a vector super computer?

(a)
Register to register
(b)
Memory to memory
(c)
Both a and b are invalid
(d)
None of the above

107

( )

( )

8.

........................is a measure of the amount of compilation involved in a software process:

(a)
Grain size
(b)
Granularity
(c)
Both a and b
(d)
neither a nor b
( )

9.

The diameter of a network is the :

(a)
Maximum shortest path between any two nodes
(b)
Minimum shortest path between any two nodes
(c)
Maximum shortest path between any two adjacent nodes
(d)
Minimum longest path between two adjacent nodes

( )

The channel width of a fat tree increase as we ascend form:

(a)
Leaves to the roots
(b)
Roots to the leaves
(c)
Width
(d)
None of the above

( )

The total number of messages the network can handle is:

(a)
Network efficiency
(b)
Network throughout
(c)
Network output
(d)
All of the above

( )

Which dynamic connection network has a low with limited bandwidth features?
(a)
Multistage interconnection
(b)
Crossbar switch
(c)
Bus system
(d)
All of the above

( )

10.

11.

12.

13.

Which dynamic connection network provides highest bandwidth and interconnection?

(a)
Crossbar networks
(b)
Multi stage networks
(c)
Bus system
(d)
None of the above
( )

14.

DOP stands for:

(a)
Dual Operating Processor
(b)
Degree of parallelism
(c)
Dual of parallelism

108

(d)

Degree of processing

( )

15.
...........................is a cascade of processing stage which are linearly connected to perform a
fixed function over a stream of data flowing from one end to the other:
(a)
Omega network
(b)
Tree Network
(c)
Linear Pipeline
(d)
Parallel Processor `
( )
16.

17.

The finest level of pipelining is called:

(a)
Macro pipelining
(b)
Micro pipelining
(c)
Man pipelining
(d)
None of the above

( )

When the addressed data/instruction is found in cache it is called:

(a)
Cache hit
(b)
Cache miss
(c)
Cache found
(d)
Cache trace

( )

18.

In proposed future bus + standard, the..................are used to declare special bus

transaction:
(a)
Command Lines
(b)
Status lines
(c)
Capability lines
(d)
Miscellaneous lines
( )

19.

The percentage of time that each pipeline stage is used over a sufficiently long series of
task initiation is the:
(a)
Pipeline throughout
(b)
Pipeline efficiency
(c)
Pipeline latency
(d)
Pipeline clock
( )

20.

The effective bandwidth available to each processor is inversely proportional

to...........contending for the bus
(a)
Number of terminal
(b)
Number of processor
(c)
Number of operating systems
(d)
All of the above
( )

21.

Flynn's classification is based on:

(a)
Memory instructions
(b)
Hardware types
(c)
Software types
(d)
Notion of instruction and data streams

22.

The size of program is determined by:

(a)
Clock Rate
(b)
Instruction Execution Rate

( )

Advanced Computer Arc.

(c)
(d)

Instruction Count
Clock Count

109

( )

23.

The............model is a special case of NUMA machine, in which the distributed main

memories are converted to caches:
(a)
UMA
(b)
Special NUMA
(c)
COMA
(d)
None of the above
( )

24.

The................network provide point to point static connections among the nodes:

(a)
Message passing
(b)
Data Passing
(c)
Both a and b
(d)
Neither a nor b

( )

MAL stands for:

(a)
Minimum allocation Latency
(b)
Minimal Average Latency
(c)
Maximum Average Latency
(d)
Maximum Allocation Latency

( )

Latencies that causes collision are called:

(a)
Minimum Allocation Latency
(b)
Minimal Average Latency
(c)
Maximum Average Latency
(d)
Maximum Allocation Latency

( )

25.

26.

27.

.....................is a pair of nodes that accounts for a disproportionately large portion of the
total network traffic.
(a)
Hit ration
(b)
Hot ratio
(c)
Hot spot
(d)
Hit spot
( )

28.

The memory hierarchy developed was based on a program behaviour is known as:
(a)
Coherence property
(b)
Locality of reference
(c)
Locality of coherence
(d)
None of the above

( )

Write through and write back are two strategies for maintaining:
(a)
Locality of reference
(b)
Collision
(c)
Coherence
(d)
All of the above

( )

29.

30.

In paging, if demanded page is not found it is declared is:

(a)
Page miss
(b)
Page fault
(c)
Both a and b

110

(d)
31.

32.

33.

34.

35.

Electronic data interchange for administration commerce and trade

LRU stands for:

(a)
Last recently used
(c)
Both a and b

(b)
(d)

Least Recently Used

Least Rarely Used

( )

( )

The performance of page replacement algorithm depends on the:

(a)
Page trace
(b)
Program behaviour
(c)
Both a and b
(d)
Neither a nor b

( )

Broad call is :
(a)
Write operation
(c)
Both a and b

( )

(b)

Read Operation
(d)
Neither a nor b

For a reservation table with n columns, the maximum forbidden latency (m):
(a)
M n
(b)
n n1
(c)
M n
(d)
n n1

( )

Intrinsic parallel computer are those that execute program in:

(a)
SIMD mode only
(b)
MIMD Mode only
(c)
SIMD and MIMD modes
(d)
None of the above

( )

36.

.......................is a shared memory system in which the access time varies with the
location of the memory word.
(a)
UMA
(b)
COMA
(c)
NUMA
(d)
All of the above
( )

37.

The topology of an interconnection network can be:

(a)
Static
(b)
Dynamic
(c)
Either a or b
(d)
None of the above

38.

39.

For inter PE data exchange network that is used be:

(a)
Static
(b)
(c)
Either a or b
(d)

( )

Dynamic
None of the above

....................networking are controlled by a global clock

(a)
Synchronous
(b)
Asynchronous
(c)
Both a and b
(d)
Neither a nor b

( )

( )

Advanced Computer Arc.

40.

111

The portion of the operating system kernel which handles the allocation and deallocation
of main memory to executing processes is called:
(a)
Memory Swapper
(b)
Memory Manager
(c)
Process Swapper
(d)
Process Manager
( )

Answer Key
1. (a)
2. (b)
11. (b)
12. (c)
21. (d)
22. (c)
31. (b)
32. (c)

3. (b)
13. (a)
23. (b)
33. (b)

4. (d)
14. (b)
24. (a)
34. (a)

5. (b)
15. (c)
25. (b)
35. (c)

6. (c)
16. (b)
26. (c)
36. (c)

______________

7. (d)
17. (a)
27. (c)
37. (c)

8. (a)
18. (a)
28. (b)
38. (b)

9. (a)
19. (c)
29. (d)
39. (a)

10. (a)
20. (b)
30. (b)
40. (d)

112

DESCRIPTIVE PART - II
Year 2007
Time allowed : 2 Hours

Maximum Marks : 30

Attempt any four questions out of the six. All questions carry 7 marks each.

Q.1

Compare the relative four cache memory organization.

(a)
Direct mapping
(b)
Fully associative
(c)
Set Associative
(d)
Sector Mapping

Q.2

Describe static and dynamic connection networks.

Q.3

Describe :
(i)
Generations of electronic computers.
(ii)
Elements of moderns computer

Q.4

Explain instruction set architecture in RISC and CISC processor.

Q.5

Explain Omega Network and Crossbar Network in detail.

Q.6

Describe the following terms associated with program partitioning and scheduling:
(a)
Grain Sizes and latency
(b)
Grain packing and scheduling

_________

Advanced Computer Arc.

113

OBJECTIVE PART- I
Year - 2006
Time allowed : One Hour

Maximum Marks : 20

The question paper contains 40 multiple choice questions with four choices and student will have
to pick the correct one. (Each carrying marks.).
1.

2.

Personal computer were appeared in:

(a)
1st generation
(b)
2nd generation
(c)
4th generation
(d)
5th generation

( )

Full form of MPP is:

(a)
Massively Pipeline Processing
(b)
Master Parallel Processor
(c)
Massive Parallel Processing
(d)
Master Pipeline Processor

( )

3.

In UMA multiprocessor model, all the processors have unequal access time to all
memory words:
(a)
True
(b)
False
( )

4.

The number of edges incident on a node is called:

(a)
Node degree
(b)
(c)
Network degree
(d)

5.

6.

7.

Node diameter
Network diameter

( )

Harold stone (1971) introduced special permutation function that was:

(a)
Multicast
(b)
Perfect Shuffle
(c)
Broadcast
(d)
Routing

( )

Switching complexity of multistage network is:

(a)
O (logk n)
(b)
(c)
O (n logk n)
(d)

( )

The finest level of pipelining is:

(a)
Macropipelining
(b)
Micropipelining
(c)
Both a and b
(d)
Neither a nor b

O( )
O (n)

( )

114

8.

Prior to 1945 computers were made up of:

(a)
Mechanical and electronic
mechanical
(c)
Electronic and electron mechanical

(b)

Mechanical

(d)

All of the above

and

electron
( )

9.

....................computing is achieved through the use of an array or processing elements

synchronized by the same controller:
(a)
MIMD
(b)
SIMD
(c)
13
(d)
None of the above
( )

10.

Barrel shiffer is a:
(a)
Static connection network
(b)
Dynamic connection network
(c)
Storage device
(d)
Switching device

11.

A greedy cycle must be:

(a)
A simple cycle
(b)
Each state appears only once
(c)
Average latency must be greater than those of other simple cycles
(a)
(c)

12.

13.

14.

15.

16.

( )

Both i and ii are true

Both ii and iii

(b)
(d)

Both i and iii

All three are true

( )

Which is the valid vector access memory scheme?

(a)
C-Access memory organization
(b)
Asynchronous
(c)
Synchronous memory organization
(d)
D-Access memory organization

( )

The full form of PRAM is:

(a)
Parallel Random Access Machine
(b)
Parallel Remote Access Machine
(c)
Powerful Random Access Memory
(d)
Partial Random Access Memory

( )

................is a measure of the amount of computation involved in a software process:

(a)
Grain size
(b)
Latency
(c)
Bisection width
(d)
Network diameter

( )

Which of the most expensive dynamic connection network?

(a)
Digital buses
(b)
Multistage
(c)
Crossbar
(d)
Can't say

( )

The accumulated rate of all state utilization determines:

(a)
Pipeline throughout
(b)
Pipeline efficiency

Advanced Computer Arc.

(c)

Bounds on MAL

115

(d)

Delay Speed

( )

17.
Processors that use multiphase clocks with a muck increase clock rate ranging from 100 to
500 MHz are:
(a)
VLIW
(b)
RISC
(c)
Memory interleaving
(d)
Super view
( )
18.

................is a process of moving blocks of information between the levels of memory

hierarchy:
(a)
Memory swapping
(b)
Memory allocation
(c)
Memory interleaving
(d)
Memory scheduling
( )

19.

Which is not a valid data routing functions?

(a)
Permutation
(b)
Perfect shuffle and exchange
(c)
Broadband
(d)
Multicast

( )

In binary tree network the bisection width would be:

(a)
1
(b)
2
(c)
N/2
(d)
{N/2}2

( )

The T3D is a :
(a)
SIMD machine
(b)
MIMD machine
(c)
Both a and b
(d)
None of the above

( )

The typical clock rate of today's CISC processor ranges from:

(a)
1 to 40 MHz
(b)
20 to 33 MHz
(c)
33 to 50 MHz
(d)
All of the above

( )

TLB is used is:

(a)
Paging
(c)
Both a and b

( )

20.

21.

22.

23.

24.

In shared cache approach:

(a)
Private caches are allowed
(b)
Shared caches are allowed
(c)
(i) is true

(b)

Segmentation
(d)
None of the above

116

(d)
25.

26.

27.

28.

None is true

( )

The effective bandwidth available to each processor is inversely:

(a)
Number of I/O devices
(b)
Number of processors
(c)
Types of I/O devices
(d)
Types of processors

( )

The main objective of scheduling evnets in a pipeline is:

(a)
to obtain shortest latency between initiations without causing cohesion
(b)
to obtain highest latency between initiations without causing collisions
(c)
To obtain shortest latency between initiations without causing collisions
(d)
All of the above

( )

In future Bus + standards, the 64 bit address lines are multiplexed with:
(a)
Lower order 64 - bit data lines
(b)
High order 64-bit data lines
(c)
Lower order 32 bit data lines
(d)
Any one

( )

The time in cycles required between the issuing of two adjacent instruction is:
(a)
Instruction issue rate
(b)
Instruction pipeline cycle
(c)
Instruction operations latency
(d)
Instruction issue latency

( )

29.

.........................refers to the process in which a resident page in main memory is repaced

by a new page transferred from the disk
(a)
Page allocation
(b)
page replacement
(c)
both a and b
(d)
Neither a nor b
( )

30.

To find out cache performance, we can use:

(a)
Program trace driven simulation
(c)
Hit ration

31.

(b)
(d)

Greedy cycles
Hit ration

If the number of links is 2NB, then this would be which kind of network:
(a)
Illiac mesh
(b)
2D Mesh
(c)
Both a and b
(d)
None of the above

( )

( )

32.

...................techniques were introduced for prefect instructions in order to overlap I/E

operations:
(a)
Sequential
(b)
Look ahead
(c)
Both a and b
(d)
None of the above
( )

33.

................uses high order bits as the module address and low order bits as the word
address within each module:

Advanced Computer Arc.

(a)
(c)
34.

117

Low order interleaving

Low order interlacing

(b)
(d)

High order interleaving

High order interlacing

( )

Printed circuit on which many connectors are used to plus in functional boards?
(a)
I/O Buses
(b)
Hierarchical Buses
(c)
Cache
(d)
Backplane Buses

( )

35.

...............................is an operation that fetches the non zero elements of sparse vector
from the memory using indicates:
(a)
Masking
(b)
Gather
(c)
Scatter
(d)
Reduction
( )

36.

The number of clock cycles between two initiations of a pipeline is:

(a)
latency cycle
(b)
Gather
(c)
Scatter
(d)
Evaluation analysis

( )

37.

....................consist of an address transfer followed by a block of I or more data to 1 or

more contiguous addresses:
(a)
Address only transfer
(b)
Packet data transfer
(c)
Evaluation cycle
(d)
All of the above
( )

38.

Three dimensions of locality property are:

(a)
Temporal, parallel and sequential
(b)
Temporal, spatial and sequential
(c)
Spatial, parallel and sequential
(d)
None of the above

( )

The motorola MC 68040 is a :

(a)
8 Hm MOS Microprocessor
(b)
.7 Hm CMOS Microprocessor
(c)
.8 Hm HCMOS Microprocessor
(d)
.7 Hm HCMOS Microprocessor

( )

39.

40.

Which is the example of blocking network?

(a)
Baseline
(c)
Delta

(b)
(d)

Omega
All of the above

( )

Answer Key
1. (c)

2. (c)

3. (b)

4. (a)

5. (b)

6. (c)

7. (b)

8. (b)

9. (c)

10. (a)

11. (a)
21. (c)
31. (b)

12. (b)
22. (c)
32. (c)

13. (a)
23. (c)
33. (b)

14. (a)
24. (c)
34. (a)

15. (c)
25. (b)
35. (a)

16. (a)
26. (d)
36. (b)

17. (c)
27. (d)
37. (d)

18. (a)
28. (d)
38. (b)

19. (c)
29. (b)
39. (c)

20. (c)
30. (c)
40. (d)

118

DESCRIPTIVE PART - II

Year 2006
Time allowed : 2 Hours
Maximum Marks : 30
Attempt any four questions out of the six. All questions carry 7 marks each.
Q.1

What do you understand by shared memory multiprocessor and distributed memory

multi computers ? Explain different models of shared memory multiprocessor.

Q.2

Explain the following terms associated with program partitioning and scheduling:
(a)
Grain sizes and latency
(b)
Grain packing and scheduling

Q.3

Explain all the factors on which the performance of interconnected network depends.
Discuss the following dynamic connection networks.
(a)
Digital buses
(b)
Omega network

Q.4

Describe:
(a)
Locality of reference
(b)
Write back caches
(c)
Write through caches

Q.5

Explain different page replacement policies with suitable example.

Q.6

What do you mean by vector processing? Explain different types of vector

instructions with example.

Advanced Computer Arc.

119

Key Terms
A

Access Time
Accumulator
Address Bus
ALU
ASCII (American
Standard Code for
Information Interchange)
ASRAM (Asynchronous
RAM)
Assembler
Associative Memory
B
Binary Operator
Bit (Binary Digit)
Boolean Algebra
Buffer
Bus
Byte
C
Cache

Time required to place read/write heads of the disk over a

particular track & sector. Rotational delay and seek time should
also be considered.
CPU register that has the outcome of operations and
occasionally the operands
System bus used to move addresses in RAM or I/O device
element of the CPU that does arithmetic and logical operations
A 7-bit standard character set that signify characters inside the
computer.
Static RAM is the one for transferring data without using the
system clock.
A system program that interpret a mnemonic assembly
language to low level language.
Memory whose location are recognized by their contents,
rather than their position
An operator that perform on two operands.
A single memory unit that has a single binary value (0 or 1)
Mathematics laws applied by the processor to do logical and
shift operations.
Memory locations used to hold input or output data. It is
required to balance difference in speed amid the CPU and the
I/O devices.
Communication path consisting of a group of lines that carry
signals, addresses, or data amid PCs elements. A bus can be
used by all computer elements.
Memory unit that embrace 8 bits.

COMA

A small, fast memory - perform like a buffer. It is used to

improve performance of CPU.
A collection of typical PC functions pooled onto one or more
integrated circuit.
Square wave with equal intervals. Used to harmonize CPU
process. Events typically happen at rising or falling rim of the
clock.
Logic circuit who's yield is a function of its input only at any
specified time. There is no storage capacity of preceding
contents of the circuit.
Cache-only memory architecture

Control Bus

System bus for transferring control signals among processor

Chipset
Clock
Combinational
(combinatorial) Circuit

120

Control Unit
CPU (Central Processing
Unit)

Computer Architecture
Computer
Organization

D
Data Bus
Decoder
Distributed Memory
DMA (Direct Memory
Access)
DRAM (Dynamic RAM)
E
EEPROM (Electronically
Erasable Programmable
Read-Only Memory)
Error-Correcting Code
Error-Detecting Code
Exclusive-OR Gate
(Function)
F
Fetch Cycle
Fixed-Point
Representation System
Flip-Flop
Floating-Point
Representation System
G

and other apparatus.

Part of the CPU accountable for calculating and coordinate
computer functions.
It is accountable for performing instructions and controlling all
other components.

the trait of a computer as noticed by the machine

language Programmer which facilitate machine language
programmer to write functionally correct, time
autonomous programs.
hardware arrangement surrounding the major operational
units, data paths, and control.
System bus for transferring data.
Combinational circuit for transferring input signal combination
on numerous input lines into one specific 2 n-output lines.
Physical memory that is alienated into module every allocated
to a processor in a multiprocessor organization.
I/O method that permit direct data swap between memory and
I/O devices without holding the processor time. CPU only
begins the I/O request and is interrupted after the transfer is
complete.
RAM put into practice using capacitors and that requires to be
sporadically re-energized
usually used in BIOS chip and can be reorganized with a
procedure known as flashing using dedicated software.
Code used for sending/receiving signals or characters used to
automatically correct errors.
Code used for sending/receiving signals or characters used to
automatically detect errors.
The logic - produce logical value True if both input values
are dissimilar else produce False.
It is used for obtaining the instruction to be carried out from
memory.
Representation of Real number in which the radix (decimal for
radix 10) is kept in a fixed place.
A memory unit that include one binary value and in which the
output signify the current state. The next state depends on
current state and the input.
Real number system in which the number is symbolize as two
distinct parts- mantissa and exponent.

Advanced Computer Arc.

Gate
GB (Gigabyte)
H
Hexadecimal Numbers
Hit Ratio
Hz (Hertz)
I
IC (Integrated Circuit /
Chip)
Instruction Format
Instruction Set
Interrupt
Interrupt Handler
ISA (Instruction Set
Architecture)

K
K (Kilo)
L
L1 (Level 1) Cache
L2 (Level 2) Cache
LAN (Local Area
Network)
Latency
Link Editor (Linker)
Local Variable
Locality (of Reference)
Principle
LRU (Least Recently
Used) Scheme
M
M (Mega)
Mantissa

121

Combinational circuit component that make an output that look

like simple Boolean function (And, OR, or NOT) of the
functional input.
= 1, 073,741, 824 bytes 1 Billion bytes
A base-16 number system that characterize 16 values (0 to 9 and
A to F). usually signify memory addresses or data.
A measure of cache efficiency = (cache hits / cache misses)
The number of clock cycles per second. by and large specified
in KHz (Kilohertz) or MHz (Megahertz).
A hardware part generally prepared of silicon that hosts dozens
to millions of transistors on a tiny region.
Instruction outline that break up instruction to fields related to
element of instruction (opcode, operands,)
Total collection of instructions used by a machine.
An exception that arrives from outside the processor.
A software program part that is run when an interrupt
happens.
An conceptual medium between the hardware and the software
of a device that include essential information to write accurate
machine-language program. It comprise the requirement of
commands, registers, memory size, instruction, ...etc.
2 10 = 1024
Cache situated contiguous to the processor. That is Primary
cache.
Cache positioned outside the processor. Known as secondary
cache.
Network link that transmit data in a small geographic region,
classically within the same building.
Wait or delay time(milliseconds).
A system program that merges separately integrate machine
language program and determine indeterminate labels. The
consequential code is in executable code form.
A variable is defined and accessed in a particular unit of a
program only.
propensity of a program to access the same set of memory
locations continually over small phase of time.
A substitute method in which the new preferred block replaces
the block that has been idle the longest time.
2 20 = 1,048,576
The part of a floating point number which, when multiply by
its base raised to the power of its exponent, provide its value.

122

MAR (Memory address

Register)
Microinstruction

Micro program

A CPU registers for keeping address of memory location being

accessed.
Low-level control instruction in which machine instruction is
used to produce control signals.
Basic CPU operation, carried out for the period of one clock
cycle.
Integrated circuits making the heart of the PC that include
ALU, general and special registers, and Control units.
Microinstruction sequence.

MIMD (Multiple
Instruction stream,
Multiple-Data stream)
MIPS (Million
Instruction Per Second)
MISD

The categorization under Flynn's nomenclature of a parallel

processor where many functional parts carry out different
function on different data.
Determine execution speed. MIPS = Number of instructions in
a program / (program execution time x 10 6 )
Multiple instruction single data stream

Miss Rate
Motherboard
Multiprocessor

Portion of memory access not found in cache.

A large printed-circuit board used to host PC components
Computer having more than one processor with common main
memory and single address room.
Programming mode that permit two or more programs to
execute interleaved by a single CPU.

Micro-operation
Microprocessor

Multiprogramming
N
Nonvolatile memory
NUMA
O
Opcode (Operation
Code)
Operand
Operating System

P
Page
Page Fault
Page Frame
Parity Bit

Memory whose data keeps integral even when the power is

turned off.
Non uniform memory access
Component of an instruction that indicate the operation and
format of an instruction.
Entity on which an operation is carry out. piece of an
instruction.
System software to control program execution, assign and deal
with resources, program tasks, control I/O operations, and
manage data.
A fixed length memory blocks that has virtual address and is
swap as a entity amid two memory types (RAM & cache or
RAM & secondary memory).
A condition that takes place when referencing a memory word
that is not in RAM. It originates interrupt & needs loading the
page ongoing the preferred word before the program can go on.
A block in RAM that can keep a page.
An extra bit attached to a word and basis for sum of all digits to
be either odd or even, depending on the type of parity (odd or

Advanced Computer Arc.

Peer-to-Peer Network
Pipeline
PRAM

123

even parity).
Two or more computers directly linked and directly share the
data and hardware resources.

The procedure employed to begin one function in every

cycle without waiting for the final outcome to be created,
or completion of formerly commenced functions.
Parallel random access machine

R
Radix (Base)
Register
S
SCSI (Small Computer
System Interface)
Seek Time
Segmentation
Sequential Circuit
Sign-Magnitude
Representation
SIMD (SingleInstruction Stream /
Single-Data Stream)
SIMD (SingleInstruction stream,
Multiple-Data stream)
(Or "data parallel")
SIMM (Single-Inline
Memory Module)
SISD (Single-Inline,
Single-Data Stream)
Snooping Cache Memory
Spatial Locality
SRAM (Static RAM)
Stack

Representation of Number system.

High-speed memory constituent exist in in the CPU used to
keep data.
A bus used as a standard for I/O devices.
Time required for the head actuator to travel the read/write
head from one track to the next.
Variable-size address mapping method in which an address is
alienated into two parts: a part number and part offset.
Logic circuit in which the next state is a function of both the
current state and the input. It works as a memory element.
Number representation used to represent binary integers. The
leftmost bit is used to represent the sign (1 for negative, 0 for
positive). rest bits keep the magnitude of the number.
Multiprocessor architecture that can do a single function on
multiple set of data.
The categorization under Flynn's nomenclature for a parallel
processor where many processing part do the same action on
different data. There is often a central controller that broadcasts
the instruction stream to all the processing elements.
Memory unit made up of DRAM chip in special packaging.
Soldered on a tiny circuit board with 30- or 72- edge connector.
Computer organization in Flynn's classification that refers to
the conservative processor.
technique for retain cache coherency in which all cache
controllers scrutinize the bus to decide whether or not they
have the preferred block.
Locality principle that states that data referencing be apt to
reference close by addresses.
RAM put into practice with flip-flops. Data keeps as long as the
power is on. No periodic revitalizing is required.
A list that is efforts on LIFO (Last-In First-Out) basis.

124

Sum-of-Product
Superscalar Pipelining
Superscalar Processor
Synchronous Timing
System Bus
T
Tag
TB (Terabytes)
TLB (Translation
Lookaside buffer
Transistor
Truth table
Throughput
U
Unary Operator
Underflow

UMA

A logical expression merge AND terms (Product) and then

applies the OR operator (Sum) on them.
A system that copies the internal computer components to
allow multiple instructions to be run in every pipeline stage.
An advanced pipelining modus operandi in which more than
one instruction can execute during one clock cycle, each on a
different pipeline stage.
Timing practice in which incidence of events on a bus are given
by the clock.
A bus used to be linked major computer components.
A field in a table having address information to discover a
memory block in which a particular word is found.
= 1, 099,511, 627, 776 bytes 1 TB
Used in virtual memory systems. A cache that keeps track of
lately used address mapping to circumvent an access to the
page table. It lists the physical address page number related
with every virtual address page number.
Electronic circuit that holds electrical voltage representing one
bit.
Table showing logic task by listing all potential input
combinations and their subsequent output values.
A measure of how many programs can be executed per second
An operator that perform on one operand only
A condition that can take place when the outcome of a floatingpoint function would be lesser in scale (closer to zero, either
positive or negative) than the smallest quantity
characterizeable. Underflow is in fact (negative) overflow of the
exponent of the floating-point quantity. It happens when a
negative exponent is too great to be characterized. That means
the number is too small to be represented.
Uniform memory access

V
Virtual Address

Virtual Memory

Volatile Memory

A memory location accessed in a system by an application

program with virtual memory such that intervening hardware
and/or software maps the virtual address to real (physical)
memory.
Address space that can be seen as addressable main memory by
the user. They are plotted by the processor into physical address
space. Usually the virtual address space is bigger than the
physical address space.
Memory that lose its contents when the power is off. Example:

Advanced Computer Arc.

W
Write Back
Write Through

125

RAM.
A cache structural design in which data is only written to main
memory when it is enforced out of the cache. Contrary to writethrough.
A cache structural design in which data is written to main
memory at the similar time as it is cached.

126

Websites
http://williamstallings.com/COA5e.html
http://iiusatech.com/murdocca/CAO/

http://courses.soe.ucsc.edu/courses/cmpe202

http://www.randelshofer.ch/fhw/gri/float.html#chapterfloatapplet
http://courses.soe.ucsc.edu/courses/cmpe110
http://booksite.mkp.com/9780123838728/

http://inst.eecs.berkeley.edu/~cs152/sp12/

http://www.randelshofer.ch/fhw/gri/float.html
http://www.britannica.com/EBchecked/topic/163278/digital-computer
http://williamstallings.com/COA/COA7e.html
http://csg.csail.mit.edu/6.823/lecnotes.html
http://www.cs.berkeley.edu/~pattrsn/252S01/
http://gs.fanshawec.ca/tlc/math270/
http://courses.cs.vt.edu/~csonline/NumberSystems/Lessons/BinaryNumb
ers/index.html

http://www.math.grin.edu/~stone/courses/fundamentals/IEEEreals.html
http://www.randelshofer.ch/fhw/gri/doubleapplet.html
http://www.intel.com/products/processor (for Intel Itanium
Processor)
http://www.ibm.com/servers/eserver/pserieslhardware/whitepaper
s/power/ppc - arch.html
http://www . i bm .com! servers/ eserver/pserics/hard ware/wh
itepapers/power/ppc arch 2.html

Advanced Computer Arc.

127

Bibliography

1.

Book Name : Computer Systems Design And Architecture

Edition: 2/e
ISBN : 9788177584837
Author : Vincent P. Heuring,Harry F. Jordan,T.G Venkatesh

2.

Book Name : Computer Architecture: Concepts And Evolution

ISBN : 9788131726143
Author : Gerrit A. Blaauw, Frederick P. Brooks
Edition : 1/e

3.

Book Name : Computer Systems Organization & Architecture

ISBN : 9788177587678
Author : John D. Carpinelli
Edition : 1/e

4.

Book Name : Essentials Of Computer Architecture

ISBN : 9788177584431
Author : Douglas E. Comer
Edition : 1/e

5.

Book Name : Principles Of Computer Organization And Assembly Language

ISBN : 9788131721711
Author : Patrick Juola
Edition : 1/e

6.

Book Name : Computer System Architecture

ISBN : 9788131700709
Author : M. Morris Mano
Edition : 3/e

7.

Book Name : Computer Organization And Architecture

ISBN : 9788131732458
Author : William Stallings
Edition : 8/e

128

8.

Book Name : Computer Architecture: A Quantitative Approach,

Author : John L. Hennessy , David A. Patterson
Edition :4th Edition

9.

Book Name : Parallel Computer Architecture: A Hardware/Software Approach

Author : David Culler and J.P. Singh with Anoop Gupta

10.

Book Name : Digital Design And Computer Organisation

ISBN : 978-81-318-0345-5
Author : D. Nasib S. Gill, J.B. Dixit
Edition : 1/e

11.

Book Name : Structured Computer Organization (5th edition)

Author : Andrew S. Tannenbaum, Prentice Hall