Lecture 02 AV-323 Computer System Architecture
Lecture 02 AV-323 Computer System Architecture
Operating Systems
Lecture No 02
August 6, 2024
Number of Slides:
1
In the Last Lecture
• Introduction and Importance of Operating Systems
• Computer System Organization
• Interrupts
• Storage structure
2
In This Lecture
• Computer System Organization
• I/O Structure
• Computer System Architecture
3
I/O Data Transfers and the CPU
• Every time a byte is to be transferred to memory,
interrupting the CPU may make the overall data
transfer very slow
• The concept of Direct Memory Access is used here
4
Direct Memory Access Structure
• Used for high-speed I/O devices able to transmit
information at close to memory speeds
• Device controller transfers blocks of data from buffer
storage directly to main memory without interrupting
the CPU for every byte
• Only one interrupt is generated in the end to signal
the completion of operation, rather than the one
interrupt per byte
5
A von Neumann architecture
6
Computer System
Architecture
Section 1.3
7
Computer System Architecture
• We shall discuss
• Single Processor Systems
• Multiprocessor Systems
• Clustered Systems
8
Single Processor System
• These refer to the computer systems having only one
general-purpose processor/core/CPU.
• We have many special-purpose processors as well.
• For example, keyboard has a microprocessor that converts
the keystrock into codes to be sent to the CPU.
• However, the microprocessor in the keyboard cannot
process all types of instructions and hence, NOT a
general-purpose processor
9
Multiprocessor System
• Multiprocessor systems growing in use and
importance
• Also known as parallel systems, tightly-coupled systems
• Advantages include:
• Increased throughput
• Economy of scale
• Increased reliability – graceful degradation or fault tolerance
• Two types:
• Asymmetric Multiprocessing – each processor is assigned
a specie task.
• Symmetric Multiprocessing – each processor performs all
tasks.
10
Symmetric Multiprocessing Architecture
11
Multiple Cores in a Single CPU
• A CPU core is the part of a CPU that can
independently process all computational tasks.
• Core is the “brain” and basic computation unit of the CPU.
• One CPU can have multiple cores.
• Each core of a CPU can perform operations
separately from the others.
• Or multiple cores may work together to perform
parallel operations on a shared set of data in the
CPU's memory cache.
12
Dual-Core Design
13
Non-Uniform Memory Access (NUMA) System
• Increasing the number of
CPUs increase contention
on the system bus.
• In NUMA, each CPU gets
its own memory to avoid
contention for the system
bus.
• NUMA is increasingly
popular due to
accommodating large
number of CPUs.
14
Clustered Systems
• A.k.a. loosely coupled systems
• Clustered systems also gather multiple CPUs.
• They differ from the multiprocessor systems in that
they are composed of two or more individual
systems/nodes—joined together.
• Each node is typically a multicore system.
• Clustered computers share storage and are closely
linked via a local-area network LAN
15
Summary
• Computer System Organization
• I/O Structure
• Computer System Architecture
• Single Processor Systems
• Multiprocessor Systems
• Clustered Systems
16