CSC413-Distributed Computing System

KWARA STATE UNIVERSITY, MALETE.

DEPARTMENT OF COMPUTER SCIENCE
2022/2023 ACADEMIC SESSION

Course Code: CSC 413

Course Title: Distributed Computing System
Course Status: 2 units
Instructor: Dr. (Mrs.) R. S. Babatunde
Course Objective: To avail the students the opportunity of understanding the concepts and
principles in designing multimedia system, using networks to enable computers to communicate
efficiently and application distribution, putting together large powerful processors connected by
high speed network. The knowledge acquired in this course is useful for Operating System and
Data Communication and Networking.

Contents
Module I
Introduction to distributed systems, characteristics of distributed systems, Design
considerations for distributed systems, Advantages and Disadvantages of Distributed Computing
System. Advantages of distributed system over centralized system. Evolution of Distributed
Computing System. Multiple Processor Connection – tightly coupled/loosely coupled systems.
Module II
Forms of Distributed Computing System – Term Paper presentation.
Models of Distributed Computing System. Challenges of Distributed Computing System/Issues
in designing Distributed Computing System. Failure Detection in Distributed Computing
System. Fault Avoidance, Tolerance, Detection and Recovery in Distributed System.

Module III
Communication in Distributed Computing System. Overview of Data Communication.
Transmission mechanism. Transmission mode. Message Transport technique.

Module IV

Client –Server Architecture. Features, Types, Advantages and disadvantages.

Expectations:- Each Student will be assigned one of the Forms of distributed computing paradigm
to research on: The overview of the form, the Key characteristics, Explanation of the fundamental
Principle behind its operation will be required and a case study scenario of problem solving with
the form. This will be in form of term paper and serve as continuous assessment test. The date
for presentation of term papers by each student will be communicated later.

1
 CSC413-Distributed Computing System

Definitions
Distributed computing is a method of computer processing in which different parts of
a program are run simultaneously on two or more computers that are communicating
with each other over a network.

Distributed computing is a computing system in which several interconnected

computers share the computing task assigned to them. Distributed computing is a
type of computing in which different components and objects of an application can be
located on different computers connected to a network. It means multiple computers
splits the work and get it done more quickly than one computer.

Distributed computing is a collection of independent computers that appear to the

users of the system as a single coherent system. It is basically a collection of processors
interconnected by a communication network in which each processor has its own local
memory and other peripherals, and the communication between processors of the system
takes place by message passing over the communication network.

Distributed computing means designing and implementing programs that run on two
or more interconnected computer systems. It is a model in which components of a
software system are shared among multiple computers to improve efficiency and
performance.

Example of Distributed Computing

 A typical distribution in a business enterprise, using a 3-tier model, user

interface processing is performed on the PC at the user's location, business processing
is done on a remote computer, and database access and processing is conducted on
another computer that provides centralized access for many business processes.
Typically, this kind of distributed computing uses the client/server communications
model.

2
 CSC413-Distributed Computing System

 An example of distributed computing is a framework in which large problems can be

divided into many small problems which are distributed to many computers. Later,
the small results are reassembled into a larger solution.
Note that: Distributed computing is a natural result of using networks to enable
computers to communicate efficiently. But distributed computing is distinct from
computer networking which refers to two or more computers interacting with each
other, but not, typically, sharing the processing of a single program. The World
Wide Web is an example of a network, but not an example of distributed
computing.

Things to consider in designing a Distributed Computing System

 Technologies and standards used to construct distributed computations, including
some which are specially designed and optimized for that purpose, such as Remote
Procedure Calls (RPC) or Remote Method Invocation (RMI) or .NET Remoting.
 Distributed computing also requires that the division of the program take into
account the different environments on which the different sections of the program
will be running. For example, two computers are likely to have different file
systems and different hardware components.
 Organizing the interaction between each computer is of prime importance. In
order to be able to use the widest possible range and types of computers, the
protocol or communication channel should not contain or use any information that
may not be understood by certain machines. Special care must also be taken that
messages are indeed delivered correctly and that invalid messages are rejected
which would otherwise bring down the system and perhaps the rest of the network.
 Another important factor is the ability to send software to another computer in a
portable way so that it may execute and interact with the existing network

The primary goal of a distributed computing system is to connect users and resources
in a transparent, open, and scalable way. Ideally this arrangement must drastically be
fault tolerant and more powerful than any combinations of stand-alone computer
systems. Therefore, the main goal of distributed computing is for device sharing, data
sharing, Communication and flexibility.

3
 CSC413-Distributed Computing System

Characteristics of Distributed System

1. Multiple Computers:
More than one physical computer, each consisting of CPUs, local memory, and possibly
stable storage, and I/O paths to connect it with the environment.

2. Interconnections and concurrent communication:

Mechanisms for node to node communication via network.

3. Shared State:
If a subset of nodes cooperate to provide a service, a shared state is maintained by these
nodes. The shared state is distributed or replicated among the participants.

Advantages of Distributed System

1. Resource sharing, provides the possibility of using available resources anywhere,
sharing of expensive resources
2. Allows openness, an open distributed system can be extended and improved
incrementally simply by publication of component interfaces and standard protocols for
accessing interfaces

3. Scalability, the ability to serve more users, provides acceptable response times with
increased amount of data and reduces cost

4. Fault tolerance, maintaining availability even when individual components fails

5. Allow heterogeneity of network and hardware, operating system, programming

languages, implementations by different developers

6. Improve modularity by reducing data coupling

7. Distribution of tasks among different machines helps to improve throughput.

Disadvantages of Distributed System

 Technical issues
1. If not planned properly, a distributed system can decrease the overall reliability of
computations, unavailability of a node can cause disruption of the other nodes.
4
 CSC413-Distributed Computing System

2. "A distributed system is one in which the failure of a computer you didn't even
know existed can render your own computer unusable. "Troubleshooting and
diagnosing problems in a distributed system can also become more difficult,
because the analysis may require connecting to remote nodes or inspecting
communication between nodes.
3. Many types of computation are not well suited for distributed environments,
typically owing to the amount of network communication or synchronization that
would be required between nodes. If bandwidth, latency, or communication
requirements are too significant, then the benefits of distributed computing may
be negated and the performance may be worse than a non-distributed
environment.
4. As distributed systems get larger, it becomes harder and harder to predict or even
understand their behavior. Part of the reason for this is that programmers have not
yet developed the kind of tools for managing complexity (like subroutines or
objects with narrow interfaces, or even simple structured programming
mechanisms like loops or if/then statements) that are standard in sequential
programming.
5. distributed systems bring with them large amounts of inherent non
determinism— unpredictable events like delays in message arrival and the
sudden failure of components

 Project-related problems
1. Distributed computing projects may generate data that is proprietary to private
industry, even though the process of generating that data involves the resources of
volunteers. This may result in controversy as private industry profits from the data
which is generated with the aid of volunteers.
2. In addition, some distributed computing projects, such as biology projects that aim
to develop thousands or millions of "candidate molecules" for solving various
medical problems, may create vast amounts of raw data. This raw data may be
useless by itself without refinement of the raw data or testing of candidate results
in real-world experiments. Such refinement and experimentation may be so

5
 CSC413-Distributed Computing System

expensive and time-consuming that it may literally take decades to sift through the
data. Until the data is refined, no benefits can be acquired from the computing
work.
3. Other projects suffer from lack of planning on behalf of their well-meaning
originators. These poorly planned projects may not generate results that are
palpable, or may not generate data that ultimately result in finished, innovative
scientific papers. Sensing that a project may not be generating useful data,
the project managers may decide to abruptly terminate the project without
definitive results, resulting in wastage of the electricity and computing resources
used in the project. Volunteers may feel disappointed and abused by such
outcomes.
4. There is an obvious opportunity cost of devoting time and energy to a project that
ultimately is useless, when that computing power could have been devoted to a
better planned distributed computing project generating useful, concrete results.
5. Some distributed computing projects may also attempt to use computers to find
solutions by number-crunching mathematical or physical models. With such
projects there is the risk that the model may not be designed well enough to
efficiently generate concrete solutions.

Advantages of Distributed Systems over Centralized Systems

• Economics: a collection of microprocessors offer a better price/performance

than mainframes. Low price/performance ratio is a cost
effective way to increase computing power.

• Speed: a distributed system may have more total computing power than a
mainframe.

• Inherent distribution: Some applications are inherently distributed.

• Reliability: If one machine crashes, the system as a whole can still survive.

• Incremental growth: Computing power can be added in small increments.

Modular expandability

• The existence of large number of personal computers, the need for people to
collaborate and share information.

6
 CSC413-Distributed Computing System

Evolution of Distributed Computing System – Read up

Multiple Processor Connection Approaches

1. Tightly coupled systems: In these systems, there is a single primary memory
(address space) that is shared by all the processors as shown in Figure 1. If any
processor writes, for example, the value 100 to the memory location x, any other
processor subsequently reading from location x will get the value 100. Therefore, in
these systems, any communication between the processors usually takes place
through the shared memory. Tightly coupled systems are referred to as parallel
processing systems,
Furthermore, in tightly coupled systems, the number of processors that can be
usefully deployed is usually small and limited by the bandwidth of the shared
memory.

Figure 1: Tightly Coupled System

2. Loosely coupled systems: In these systems, the processors do not share memory,
and each processor has its own local memory as shown in Figure 2. If a processor
writes the value 100 to the memory location x, this write operation will only change
the contents of its local memory and will not affect the contents of the memory. In
these systems, all physical communication between the processors is done by passing
messages across the network that interconnects the processors. Loosely coupled
systems are referred to as distributed computing systems, or simply distributed
systems.

7
 CSC413-Distributed Computing System

Figure 2: Loosely Coupled System

In contrast to the tightly coupled systems, the processor of distributed computing systems
can be located far from each other to cover a wider geographical area. In distributed
computing systems processors are more freely expandable and can have an almost unlimited
number of processors.

In short, a distributed computing system is basically a collection of processors

interconnected by a communication network in which each processor has its own local
memory and other peripherals, and the communication between any two processors of the
system takes place by message passing over the communication network.
Together, a processor and its resources are usually referred to as a node or site or machine
of the distributed computing system.

Forms of Distributed Computing System – Term Paper Presentation

Models of Distributed Computing System

Various models are used for building distributed computing systems. These models can be
broadly classified into five categories – minicomputer, workstation, workstation-server,
Processor-pool, and hybrid.
1. Minicomputer Model:
The minicomputer model is a simple extension of the centralized time sharing system
as shown in Figure 3. A distributed computing system based on this model consists of a few
minicomputers (they may be large supercomputers as well) interconnected by a

8
 CSC413-Distributed Computing System

communication network. Each minicomputer usually has multiple users simultaneously

logged on to it. For this, several interactive terminals are connected to each minicomputer.
Each user is logged on to one specific minicomputer, with remote access to other
minicomputers. The network allows a user to access remote resources that are available on
some machine other than the one on to which the user is currently logged.
The minicomputer model may be used when resource sharing (Such as sharing of
information databases of different types, with each type of database located on a different
machine) with remote users is desired. The early ARPAnet is an example of a distributed
computing system based on the minicomputer model.

Figure 3: A distributed Computing System based on the minicomputer model

2. Workstation Model:

A distributed computing system based on the workstation model consists of several

workstations interconnected by a communication network as shown in Figure 4. A
company’s office or a university department may have several workstations scattered
throughout a building or campus, each workstation equipped with its own disk and serving
as a single-user computer.
It has been often found that in such an environment, at any one time (especially at night), a
significant proportion of the workstations are idle (not being used), resulting in the waste
of large amounts of CPU time. Therefore, the idea of the workstation model is to interconnect
all these workstations by a high speed LAN so that idle workstations may be used to process

9
 CSC413-Distributed Computing System

jobs of users who are logged onto other workstations and do not have sufficient processing
power at their own workstations to get their jobs processed efficiently. In this model, a user
logs onto one of the workstations called his or her “home” workstation and submits jobs for
execution. When the system finds that the user’s workstation does not have sufficient
processing power for executing the processes of the submitted jobs efficiently, it transfers
one or more of the process from the user’s workstation to some other workstation that is
currently idle and gets the process executed there, and finally the result of execution is
returned to the user’s workstation.

Figure 4: A distributed computing system based on the workstation model

The workstation model is a network of personal workstations, each with its own disk and a
local file system. A workstation with its own local disk is usually called a diskful workstation
and a workstation without a local disk is called a diskless workstation. With the proliferation
of high-speed networks, diskless workstations have become more popular in network
environments than diskful workstations, making the workstation-server model more
popular than the workstation model for building distributed computing systems.
3. Workstation Server Model
A distributed computing system based on the workstation server model consists of a few
minicomputers and several workstations (most of which are diskless, but a few of which may
be diskful) interconnected by a communication network.

Normal computation activities required by the user’s processes are preformed at the user’s
home workstation, but requests for services provided by special servers (such as a file server

10
 CSC413-Distributed Computing System

or a database server) are sent to a server providing that type of service that performs the
user’s requested activity and returns the result of request processing to the user’s
workstation.

Therefore, in this model, the user’s processes need not be migrated to the server machines
for getting the work done by those machines.
For better overall system performance, the local disk of a diskful workstation is normally
used for such purposes as storage of temporary files, storage of unshared files, storage of
shared files that are rarely changed, paging activity in virtual-memory management, and
changing of remotely accessed data.

As compared to the workstation model, the workstation–server model has several

advantages:

1. In general, it is much cheaper to use a few minicomputers equipped with large, fast disks
that are accessed over the network than a large number of diskful workstations, with each
workstation having a small, slow disk.

2. Diskless workstations are also preferred to diskful workstations from a system

maintenance point of view. Backup and hardware maintenance are easier to perform with a
few large disks than with many small disks scattered all over a building or campus.
Furthermore, installing new releases of software (Such as a file server with new
functionalities) is easier when the software is to be installed on a few file server machines
than on every workstations.

3. In the workstation server model, since all files are managed by the file servers, user have
the flexibility to use any workstation and access the files in the same manner irrespective of
which workstation the user is currently logged on. Note that this is not true with the
workstation model, in which each workstation has its local file system, because different
mechanisms are needed to access local and remote files.

4. In the workstation server model, the request response protocol is mainly used to access
the services of the server machines. Therefore, unlike the workstation model, this model
does not need a process migration facility, which is difficult to implement.

11
 CSC413-Distributed Computing System

The request response protocol is known as the client-server model of communication. In

this model, a client process (which in this case resides on a workstation) sends a request to
a server process (Which in this case resides on a minicomputer) for getting some service
such as a block of a file. The server executes the request and sends back a reply to the client
that contains the result of request processing.
The client-server model provides an effective general –purpose approach to the sharing of
information and resources in distributed computing systems.

5. A user has guaranteed response time because workstations are not used for executing
remote processes. However, the model does not utilize the processing capability of idle
workstations.

Figure 5: Distributed Computing System based on the workstation server model

4. Processor Pool Model:

The processor – pool model is based on the observation that most of the time a user does not
need any computing power but once in a while he or she may need a very large amount of
computing power for a short time. (e.g., when recompiling a program consisting of a large
number of files after changing a basic shared declaration). Therefore, unlike the workstation
– server model in which a processor is allocated to each user, in the processor-pool model
the processors are pooled together to be shared by the users as needed.

12
 CSC413-Distributed Computing System

The pool of processors consists of a large number of microcomputers and minicomputers

attached to the network. Each processor in the pool has its own memory to load and run a
system program or an application program of the distributed computing system.

As shown in Figure 6, in the pure processor-pool model, the processors in the pool have no
terminals attached directly to them, and users access the system from terminals that are
attached to the network via special devices. These terminals are either small diskless
workstations or graphic terminals, such as X terminals. A special server (Called a run server)
manages and allocates the processors in the pool to different users on a demand basis. When
a user submits a job for computation, an appropriate number of processors are temporarily
assigned to his or her job by the run server. For example, if the user’s computation job is the
compilation of a program having n segments, in which each of the
segments can be complied independently to produce separate relocatable object files, n
processors from the pool can be allocated to this job to compile all the n segments in parallel.
When the computation is completed, the processors are returned to the pool for use by other
users. In the processor-pool model there is no concept of a home machine. That is, a user
does not log onto a particular machine but to the system as a whole.

Figure 6: A distributed computing system based on the processor-pool model

13
 CSC413-Distributed Computing System

5. Hybrid Model:

Out of the four models described above, the workstation server model, is the most
widely used model for building distributed computing systems. This is because a large
number of computer users only perform simple interactive tasks such as editing jobs,
sending electronic mails, and executing small programs. The workstation-server model is
ideal for such simple usage.
However, in a working environment that has groups of users who often perform jobs
needing massive computation, the processor-pool model is more attractive and suitable. The
advantages of both the workstation-server and processor-pool models results in a hybrid
model used to build a distributed computing system.
The hybrid model is based on the workstation-server model but with the addition of
a pool of processors. The processors in the pool can be allocated dynamically for
computations that are too large for workstations or that requires several computers
concurrently for efficient execution. In addition to efficient execution of computation-
intensive jobs, the hybrid model gives guaranteed response to interactive jobs by allowing
them to be processed on local workstations of the users.
However, the hybrid model is more expensive to implement than the workstation – server
model or the processor-pool model.

CHALLENGES/DESIGN ISSUES OF DISTRIBUTED SYSTEM

In general, designing a distributed operating system is more difficult than designing
a centralized operating system for several reasons. In the design of a centralized operating
system, it is assumed that the operating system has access to complete and accurate
information about the environment in which it is functioning. For example, a centralized
operating system can request status information, being assured that the interrogated
component will not charge state while awaiting a decision based on that status information,
since only the single operating system asking the question may give commands.
However, a distributed operating system must be designed with the assumption that
complete information about the system environment will never be available.

14
 CSC413-Distributed Computing System

A. Transparency
A distributed system that is able to present itself to user and application as if it were only a
single computer system is said to be transparent. There are eight types of transparencies in
a distributed system:
1) Access Transparency: It hides differences in data representation and how a resource is
accessed by a user. Example, a distributed system may have a computer system that runs
different operating systems, each having their own file naming conventions. Differences in
naming conventions as well as how files can be manipulated should be hidden from the users
and applications.

2) Location Transparency: Hides where exactly the resource is located physically. Example,
by assigning logical names to resources like yahoo.com, one cannot get an idea of the location
of the web page’s main server.

3) Migration Transparency: Distributed system in which resources can be moved without

affecting how the resource can be accessed are said to provide migration transparency. It
hides that the resource may move from one location to another.

4) Relocation Transparency: this transparency deals with the fact that resources can be
relocated while it is being accessed without the user who is using the application to know
anything.
Example: using a Wi-Fi system on laptops while moving from place to place without getting
disconnected.

5) Replication Transparency: Hides the fact that multiple copies of a resource could exist
simultaneously. To hide replication, it is essential that the replicas have the same name.
Consequently, as system that supports replication should also support location
transparency.

15
 CSC413-Distributed Computing System

6) Concurrency Transparency: It hides the fact that the resource may be shared by several
competitive users. Example, two independent users may each have stored their file on the
same server and may be accessing the same table in a shared database. In such cases, it is
important that each user doesn’t notice that the others are making use of the same resource.

7) Failure Transparency: Hides failure and recovery of the resources. It is the most difficult
task of a distributed system and is even impossible when certain apparently realistic
assumptions are made. Example: A user cannot distinguish between a very slow or dead
resource. Same error message come when a server is down or when the network is
overloaded of when the connection from the client side is lost. So here, the user is unable to
understand what has to be done, either the user should wait for the network to clear up, or
try again later when the server is working again.

8) Persistence Transparency: It hides if the resource is in memory or disk. Example, Object

oriented database provides facilities for directly invoking methods on storage objects. First
the database server copies the object states from the disk i.e. main memory performs the
operation and writes the state back to the disk. The user does not know that the server is
moving between primary and secondary memory.

B Reliability

In general, distributed systems are expected to be more reliable than centralized

systems due to the existence of multiple instances of resources. However, the existence of
multiple instances of the resources alone cannot increase the system’s reliability. Rather, the
distributed operating system, which manages these resources must be designed properly to
increase the system’s reliability by taking full advantage of this characteristic feature of a
distributed system.
Making a distributed system reliable is very important. The failure of a distributed system
can result in anything from easily repairable errors to catastrophic meltdowns. A reliable
distributed system is designed to be as fault tolerant as possible. Fault tolerance deals with
making the system function in the presence of faults. Faults can occur in any of the

16
 CSC413-Distributed Computing System

components of a distributed system. Some types of faults in a system and some of their
solutions include:

Component Faults
There are three types of component faults: transient faults, intermittent faults and
permanent faults.
- A transient fault occurs once and then disappears. If the operation is repeated then
the system will behave normally.
- An intermittent fault arises, then goes away, rises again and so on. A common cause
of an intermittent fault is a loose contact on a connector. These faults are very
annoying and hard to fix because of their sporadic nature.
- Permanent faults caused by faulty components. The system will not work until the
component is replaced. Burnt out chips, software bugs and processor failure are all
examples of permanent faults.

Processor Faults
A special type of component is the processor, and it can fail in three ways, fail-silent,
Byzantine and slowdown. All lead to a different kind of failure.
- A fail-silent, also known as fail-stop fault occurs when a processor stops
functioning. It no longer accepts input and outputs nothing, except perhaps to say it
is no longer functioning.
- Byzantine faults occurs when a faulty processor continues to run, giving wrong
answers and maybe working with other faulty processors to give the impression that
they are working correctly. Compared with fail-silent faults, Byzantine faults are hard
to diagnose and more difficult to deal with.
- A slowdown fault occurs when a certain processor executes very slowly. That
processor might be labeled as failed by the other processors. However the “failed”
processor may return to its normal speed and begin to issue orders, causing problems
within the distributed systems.

17
 CSC413-Distributed Computing System

Network Failures
Network failures keep processors from communicating with each other. We will look at the
failures resulting in total loss of communication along parts of the network. Two problems
arise from this: one-way links and network partitions.
- One-way links cause problems similar to processor slowdown. For example,
processor A can send messages to processor B but cannot receive messages from
B. Processor C can talk to both A and B. So each processor has a different idea of which
processors have failed. Processor A might think that B has failed since it does not
receive any messages from it. Processor C thinks both A and B are working properly
since it can send and receive messages from both.
- Network partitions occur when a line connecting two sections of a network fail. So
processors A and B can communicate with each other but cannot communicate with
processors C and D and visa versa. Let us say processor A updates a file and processor
C updates the same file but in a different way. When the partition is fixed, the file will
not be consistent among all processors. It is not clear to the processors how to make
the data consistent again.

C. Flexibility

Another important issue in the design of distributed operating systems is flexibility. Flexibility
is the most important features for open distributed systems. The design of a distributed
operating system should be flexible due to the following reasons:

1. Ease of modification : From the experience of system designers, it has been found that
some parts of the design often need to be replaced / modified either because some bug is
detected in the design or because the design is no longer suitable for the changed system
environment or new-user requirements. Therefore, it should be easy to incorporate
changes in the system in a user-transparent manner or with minimum interruption
caused to the users.
2. Ease of enhancement : In every system, new functionalities have to be added from time
to time it more powerful and easy to use. Therefore, it should be easy to add new services
18
 CSC413-Distributed Computing System

to the system. Furthermore, if a group of users do not like the style in which a particular
service is provided by the operating system, they should have the flexibility to add and
use their own service that works in the style with which the users of that group are more
familiar and feel more comfortable. The most important design factor that influences the
flexibility of a distributed operating system is the model used for designing its kernel. The
kernel of an operating system is its central controlling part that provides basic system
facilities. It operates in a separate address space that is inaccessible to user processes. It
is the only part of an operating system that a user cannot replace or modify.
The two commonly used models for kernel design in distributed operating systems are the
monolithic kernel and the microkernel.
- In the monolithic kernel model, most operating system
services such as process management, memory management, device management, file
management, name management, and inter-process communication are provided by the kernel.
As a result, the kernel has a large, monolithic structure. Many distributed operating systems that
are extensions or limitations of the UNIX operating system use the monolithic kernel model. This
is mainly because UNIX itself has a large, monolithic kernel.
- On the other hand, in the microkernel model, the main goal is to keep the kernel as small
as possible. Therefore, in this model, the kernel is a very small nucleus of software that provides
only the minimal facilities necessary for implementing additional operating system services. The
only services provided by the kernel in this model are inter-process communication low level
device management, a limited amount of low-level process management and some memory
management. All other operating system services, such as file management, name management,
additional process, and memory management activities and much system call handling are
implemented as user-level server processes. Each server process has its own address space and
can be programmed separately.

As compared to the monolithic kernel model, the microkernel model has several advantages. In
the monolithic kernel model, the large size of the kernel reduces the overall flexibility and
configurability of the resulting operating system. On the other hand, the resulting operating
system of the microkernel model is highly modular in nature. Due to this characteristic feature,
the operating system of the microkernel model is easy to design, implement, and install.
Moreover, since most of the services are implemented as user-level server processes, it is also

19
 CSC413-Distributed Computing System

easy to modify the design or add new services. In spite of its potential performance cost, the
microkernel model is being preferred for the design of modern distributed operating systems.

D. Performance

If a distributed system is to be used its performance must be at least as good as a

centralized system. That is, when a particular application is run on a distributed system, its
overall performance should be better than or at least equal to that of running the same
application on a single processor system.
However, to achieve his goal, it is important that the various components of the operating
system of a distributed system be designed properly; otherwise, the overall performance of the
distributed system may turn out to be worse than a centralized system. Some design principles
considered useful for better performance are as follows:

1. Batch if possible: Batching often helps in improving performance greatly. For example,
transfer of data across the network in large chunks rather than as individual pages is much more
efficient. Similarly, piggybacking of acknowledgement of previous messages with the next
message during a series of messages exchanged between two communicating entities also
improves performance.

2. Cache whenever possible: Caching of data at clients’ sites frequently improves overall
system performance because it makes data available wherever it is being currently used, thus
saving a large amount of computing time and network bandwidth. In addition, caching reduces
contention on centralized resources.

3. Minimize copying of data: Data copying overhead (e.g. moving data in and out of buffers)
involves a substantial CPU cost of many operations. For example, while being transferred from
its sender to its receiver, a message data may take the following path on the sending side:
(a) From sender’s stack to its message buffer
(b) From the message buffer in the sender’s address space to the message buffer in the kernel’s
address space
(c) Finally, from the kernel to the network interface board.

20
 CSC413-Distributed Computing System

On the receiving side, the data probably takes a similar path in the reverse direction. Therefore,
in this case, a total of six copy operations are involved in the message transfer operation.
Similarly, in several systems, the data copying overhead is also large for read and write
operations on block I/O devices.
Therefore, for better performance, it is desirable to avoid copying of data, although this is not
always simple to achieve. Making optimal use of memory management often helps in eliminating
much data movement between the kernel, block I/O devices, clients, and servers.

4. Minimize network traffic: System performance may also be improved by reducing internode
communication costs. For example, accesses to remote resources require communication,
possibly through intermediate nodes. Therefore, migrating a process closer to the resources it is
using most heavily may be helpful in reducing network traffic in the system if the decreased cost
of accessing its favorite resource offsets the possible increased post of accessing its less favored
ones. Another way to reduce network traffic is to use the process migration facility to cluster two
or more processes that frequently communicate with each other on the same node of the system.
Avoiding the collection of global state information for making some decision also helps in
reducing network traffic.

5. Take advantage of fine-grain parallelism for multiprocessing: Performance can also be

improved by taking advantage of parallelism for multiprocessing. For example, threads are often
used for structuring server processes. Servers structured as a group of threads can operate
efficiently, because they can simultaneously service requests from several clients.

E. Scalability

Scalability refers to the capability of a system to adapt to increased service load. It is inevitable
that a distributed system will grow with time since it is very common to add new machines or
an entire subnetwork to the system to take care of increased workload or organizational changes
in a company. Therefore, a distributed operating system should be designed to easily cope with
the growth of nodes and users in the system. That is, such growth should not cause serious
disruption of service or significant loss of performance to users.

21
 CSC413-Distributed Computing System

In the design of a distributed operating system, use of centralized entities such as a single central
file server or a single database for the entire system makes the distributed system non-scalable

Security in Distributed Computing System

In order that the users can trust the system and rely on it, the various resources of a
computer system must be protected against destruction and unauthorized access. Enforcing
security in a distributed system is more difficult than in a centralized system because of the lack
of a single point of control and the use of insecure networks for data communication. In a
centralized system, all users are authenticated by the system at login time, and the system can
easily check whether a user is authorized to perform the requested operation on an accessed
resource. In a distributed system, however, since the client – server model is often used for
requesting and providing services, when a client sends a request message to a server, the server
must have some way of knowing who the client is. This is not so simple as it might appear
because any client identification field in the message cannot be trusted. This is because an
intruder (a person or program trying to obtain unauthorized access to system resources) may
pretend to be an authorized client or may change the message contents during transmission.

Therefore, as compared to a centralized system, enforcement of security in a

distributed system has the following additional requirements:

1. It should be possible for the sender of a message to know that the message was received by
the intended receiver.
2. It should be possible for the receiver of a message to know that the message was sent by the
genuine sender.
3. It should be possible for both the sender and receiver of a message to be guaranteed that the
contents of the message were not changed while it was in transfer.

Cryptography is the only known practical method for dealing with these security aspects of a
distributed system. In this method comprehension of private information is prevented by
encrypting the information, which can then be decrypted only by authorized users.

22
 CSC413-Distributed Computing System

Failure Detection in Distributed Computing System

Failure detector is an application that is responsible for detection of node failures or

crashes in a distributed system. It provides hints about the operational status of other processes.
The detection of process failures is a crucial issue which system designers have to cope with in
order to build fault tolerant distributed platforms.

Many distributed systems must handle crash failures, such as application crashes, operating
system crashes, device driver crashes, application deadlocks, and hardware failures. A common
way to handle crashes involves two steps: (1) Detect the failure; and (2) Recover, by restarting.

A common way for a distributed system to tolerate crashes is to explicitly detect them and then
recover from them. Interestingly, detection can take much longer than recovery, as a result of
many advances in recovery techniques, making failure detection the dominant factor in these
systems’ unavailability when a crash occurs. A failure detector is a service that reports the status
of a remote process as UP or DOWN. A failure detector should ideally have three properties.

i. First, it should be a reliable failure detector (RFD): when a process is up, it is reported
as UP, and when it crashes, it is reported as DOWN after a while.

ii. Second, the failure detector should be fast: the time taken to report DOWN, known as
the detection time, should be short (less than a second), so as not to delay recovery.

iii. Third, the failure detector should cause little disruption.

The above properties are in tension with each other and with other desired properties. For
instance, a short detection time based on timeouts would compromise reliability, since the
detector would report as DOWN a process that is UP. As an alternative, a detector could ensure
reliability and a short detection time by killing processes at the slightest provocation, but that
would be disruptive. Also, short detection times often require probing the target incessantly,
which is costly. Another challenge is comprehensiveness: how can the detector maximize its
coverage of failures?

Synchronous Vs Asynchronous Communication

A distributed system is synchronous if the time to execute each step of a process has an
upper and lower bound. Each message transmitted over a channel is received within a known

23
 CSC413-Distributed Computing System

bounded delay. Each process has a local clock whose drift rate from real time has a known
bound.

In synchronous message-passing, every process sends out messages at time t that are delivered
at time t+1, at which point more messages are sent out that are delivered at time t + 2, and so on:
the whole system runs in lockstep, marching forward in perfect synchrony. Such systems are
difficult to build when the components become too numerous or too widely dispersed, but they
are often easier to analyze. Timeouts can be used to detect failures.

A distributed system is asynchronous if each step of a process can take an arbitrary time.
i.e. message delivery time and clock drift rates are arbitrary. In this case it is difficult to build a
failure detector or almost impossible to detect failure. Messages are delivered eventually after
some unknown delay. Unfortunately, it is impossible to distinguish with certainty a crashed
process from a very slow process in a purely asynchronous distributed system.

Fault Avoidance in Distributed Computing System

A fault is a mechanical or algorithmic defect that may generate an error. A fault in a
system causes system failure. Fault avoidance deals with designing the components of the
system in such a way that the occurrence of faults in minimized. Conservative design practice
such as using high reliability components are often employed for improving the system’s
reliability based on the idea of fault avoidance. Although a distributed operating system often
has little or no role to play in improving the fault avoidance capability of a hardware
component, the designers of the various software components of the distributed
operating system must test them thoroughly to make these components highly reliable.

Fault Tolerance in Distributed Computing System

Fault tolerance is the ability of a system to continue functioning in the event of partial
system failure. The performance of the system might be degraded due to partial failure, but
otherwise the system functions properly. Some of the important concepts that may be used
to improve the fault tolerance ability of a distributed operating system are as follows:
1. Redundancy techniques: The basic idea behind redundancy techniques is to avoid single
points of failure by replicating critical hardware and software components, so that if one of

24
 CSC413-Distributed Computing System

them fails, the others can be used to continue. Obviously, having two or more copies of a
critical component makes it possible, at least in principle, to continue operations in spite of
occasional partial failures. For example, a critical process can be simultaneously executed on
two nodes so that if one of the two nodes fails, the execution of the process can be completed
at the other node. Similarly, a critical file may be replicated on two or more storage devices
for better reliability. Notice that with redundancy techniques additional system overhead is
needed to maintain two or more copies of a replicated resource and to keep all the copies of
a resource consistent. For example, if a file is replicated on two or more nodes of a distributed
system, additional disk storage space is required and for correct functioning, it is often
necessary that all the copies of the file are mutually consistent. In general, the larger is the
number of copies kept, the better is the reliability
but the incurred overhead involved.
2. Distributed control: For better reliability, many of the particular algorithms or protocols
used in a distributed operating system must employ a distributed control mechanism to
avoid single points of failure. For example, a highly available distributed file
system should have multiple and independent file servers controlling multiple and
independent storage devices. In addition to file servers, a distributed control technique could
also be used for name servers, scheduling algorithms, and other executive control functions.

Fault Detection and Recovery in Distributed Computing System

Fault detection and recovery deals with the use of hardware and software
mechanisms to determine the occurrence of a failure and then to correct the system to a state
acceptable for continued operation. Some of the commonly used techniques for
implementing this method in a distributed operating system are as follows:

1. Atomic transactions: An atomic transaction is a computation consisting of a collection

of operation that take place indivisibly in the presence of failures and concurrent
computations. That is, either all of the operations are performed successfully or none of the
operations is performed. Other processes executing concurrently cannot modify or observe
intermediate states of the computation.

25
 CSC413-Distributed Computing System

Transactions help to preserve the consistency of a set of shared data objects (e.g. files) in the
face of failures and concurrent access. They make crash recovery much easier, because
transactions can only end in two states : Either all the operations of the transaction are
performed or none of the operations of the transaction is performed.

In a system with transaction facility, if a process halts unexpectedly due to a hardware error
before a transaction is completed, the system subsequently restores any data objects that
were undergoing modification to their original states. Notice that if a system does not
support a transaction mechanism, unexpected failure of a process during the processing of
an operation may leave the data objects that were undergoing modification in an
inconsistent state. Therefore, without transaction facility, it may be difficult or even
impossible in some cases to roll back (recover) the data objects from their current
inconsistent states to their original states.

2. Stateless servers: The client-server model is frequently used in distributed systems to

service user requests. In this model, a server may be implemented by using any one of the
following two service paradigms – stateful or stateless. The two paradigms are distinguished
by one aspect of the client – server relationship, whether or not the history of the serviced
requests between a client and a server affects the execution of the next service request.
The stateful approach does depend on the history of the serviced requests, but the stateless
approach does not depend on it. Stateless servers have a distinct advantage over stateful
servers in the event of a failure. That is, the stateless service paradigm makes crash recovery
very easy because no client state information is maintained by the server.
On the other hand, the stateful service paradigm requires complex crash recovery
procedures. Both the client and server need to reliably detect crashes. The server needs to
detect client crashes so that it can discard any state it is holding for the client, and the client
must detect server crashes so that it can perform necessary error – handling activities.

Although stateful service becomes necessary in some cases, to simplify the failure detection
and recovery actions, the stateless service paradigm must be used, wherever possible.

26
 CSC413-Distributed Computing System

3. Acknowledgments and timeout-based retransmission of messages. In a distributed

system, events such as a node crash or a communication link failure may interrupt a
communication that was in progress between two processes, resulting in the loss of a
message. Therefore, a reliable interprocess communication mechanism must have ways to
detect lost messages so that they can be retransmitted. Handling of lost messages usually
involves return of acknowledgment messages and retransmissions on the basis of timeouts.
That is, the receiver must return an acknowledgment message for every message received,
and if the sender does not receive any acknowledgement for a message within a fixed timeout
period, it assumes that the message was lost and retransmits the message.

Communication in Distributed Computing

Overview of Data Communications
Data communications is the exchange of data between two devices via some form of
transmission medium (wired/wireless). For data communications to occur, the
communicating devices must be part of a communication system made up of a
combination of hardware (physical equipment) and software (programs). The
effectiveness of a data communications system depends on four fundamental
characteristics: delivery, accuracy, timeliness, and jitter.

1. Delivery: The system must deliver data to the correct destination. Data must be
received by the intended device or user and only by that device or user.

2. Accuracy: The system must deliver the data accurately. Data that have been
altered in transmission and left uncorrected are unusable.

3. Timeliness: The system must deliver data in a timely manner. Data delivered late
are useless. In the case of video and audio, timely delivery means delivering data
as they are produced, in the same order that they are produced, and without
significant delay. This kind of delivery is called real-time transmission.

27
 CSC413-Distributed Computing System

4. Jitter: Jitter refers to the variation in the packet arrival time. It is the uneven delay
in the delivery of packets. For example, let us assume that video packets are sent
every 30ms. If some of the packets arrive with 30-ms delay and others with 40-ms
delay, an uneven quality in the video is the result.

Components of Data communication system

A data communications system has five components.

1. Message. The message is the information (data) to be communicated. Popular

forms of information include text, numbers, pictures, audio, and video.

2. Sender. The sender is the device that sends the data message. It can be a
computer, workstation, telephone handset, video camera, and so on.

3. Receiver. The receiver is the device that receives the message. It can be a
computer, workstation, telephone handset, television, and so on.

4. Transmission medium. The transmission medium is the path by which a

message travels from sender to receiver. Some examples of transmission media
include twisted-pair wire, coaxial cable, fiber optic cable, and radio waves.

5. Protocol. A protocol is a set of rules that govern data communications. It sends

an agreement between the communicating devices. Without a protocol, devices
may be connected but not communicating, just as a person speaking French cannot
be understood by a person who speaks only Japanese.

Figure 7: Data Communication Components

28
 CSC413-Distributed Computing System

Types of transmission mechanism

1. Connection oriented

Connection-oriented communication is a network communication mode where a

communication session is established before any useful data can be transferred, and
where a stream of data is delivered in the same order as it was sent. Connection-oriented
communication uses packet-mode virtual circuit connection in which either a
transport layer virtual circuit protocol such as the TCP protocol, allowing data to be
delivered in order, or it may be a data link layer or network layer switching mode, where
all data packets belonging to the same traffic stream are delivered over the
same path. Connection-oriented protocol services are often reliable network services,
that provide acknowledgment after successful delivery, and automatic repeat request
functions in case of missing data or detected bit-errors Accurate and timely transmission
of data is guaranteed since – due to acknowledgement and retransmission. Flow control
and error correction functions incur huge overhead data and hence add to delay.

2. Connectionless network

The alternative to connection-oriented transmission is connectionless communication, for

example the datagram mode communication used by the IP and UDP protocols, where data
may be delivered out of order, since different packets are routed independently, and
may be delivered over different paths. It does not support flow control and error correction
functions but offer framing as well as detection of errors. It provides only datagram services. The
UDP has less overhead because no sequencing or acknowledgement are used

Types of Network transmission mode

A network is two or more devices connected through links. A link is a communications

pathway that transfers data from one device to another. For visualization purposes, it is
simplest to imagine any link as a line drawn between two points. For communication to
occur, two devices must be connected in some way to the same link at the same time.
There are two possible types of connections:

29
 CSC413-Distributed Computing System

1. point-to-point/unicast. A point-to-point connection provides a dedicated link

between two devices. The entire capacity of the link is reserved for transmission
between those two devices. In this mode, individual pair of machine on the network
may have several connections in between them. Most point-to-point connections
use an actual length of wire or cable to connect the two ends, but other options,
such as microwave or satellite links, are also possible. When you change television
channels by infrared remote control, you are establishing a point-to-point
connection between the remote control and the television's control system.

2. Multipoint/Broadcast. A multipoint also called a broadcast connection is one

in which more than two specific devices share a single link. It has a single
communication channel that is shared by all the machines on the network. In a
multipoint environment, the capacity of the channel is shared, either spatially or
temporally. If several devices can use the link simultaneously; it is a spatially
shared connection. If users must take turns, it is a timeshared connection. E.g
KWTV, DSTV– one point reaching out to many. A packet sent by any of the
machines on the network are received by all other machines on the network. On
receiving the packet, each machine examine it for ownership. The machine
processes the packet if it is intended for it. Otherwise it is ignored. Multicasting is
a subset of broadcast network which support addressing a subset of the computers
on the network.

Message Transport Techniques

There are three types of message transport techniques namely; circuit switched, message
switched and packet switched.
1. Circuit Switched Network. It transmits a message by providing a complete
path of transmission link right from the source of the message to the destination
node. A path is set up by a special signaling sent by the source and acknowledged
by the destination.
In this technique, data is transmitted progressively along the path without no
intermediate store-and-forward delay. The entire path is allocated to this

30
 CSC413-Distributed Computing System

transmission until the source node releases this path. There is the requirement of
a dedicated path between the calling and the called parties. E.g. Telephone System.
2. Message Switched Network. It routes a message through intermediate nodes
which buffer the message data. Message is stored by an intermediate node and
forwarded to the next node upon allocation of buffer for the data at the next node.
It is called store and forward network.
In a message-switched network, there is no direct connection between the source
and destination nodes. In such networks, the intermediary nodes (switches) have
the responsibility of conveying the received message from one node to another in
the network. Therefore, each intermediary node within the network must store all
messages before retransmitting them one at a time as proper resource become
available. E.g. Telegraph. Message switching offers a number of attractive benefits,
including efficient usage of network resources, traffic regulation, and message
storage when messages cannot be delivered. There is need for large storage space
to store data in nodes
3. Packet Switched Network. This is similar to message switched network. The
only difference is that the message to be sent is broken up into fixed size segments
called packets. The packets traverse independently in the network until they reach
the destination node where they are reassembled.

Communication Protocols

In computer networks, communication occurs between two entities capable of sending

or receiving information. However, two entities do not simply send bit streams to each
other and expect to be understood. For communicate to occur, the entities must agree on
a protocol. A protocol is a set of rules that govern communications. A protocol defines
what is communicated, how it is communicate, when it is communicated. The key
elements of a protocol are syntax, semantics, and timing.

1. Syntax: The term syntax refers to the structure or format of the data, and order
in which they are presented. For example, a simple protocol might expect first 8

31
 CSC413-Distributed Computing System

bits of data to be the address of the sender, the second 8 bits to be the address of
the receiver, and the rest of the stream to be the message itself.

2. Semantics: The word semantics refers to the meaning of each section or

how is a particular pattern to be interpreted, and what action is to be taken
on that interpretation? For example, does an address identify the route to be or the
final destination of the message?

3. Timing: The term timing refers to when data should be sent and how fast
they can be sent. For example, if a sender produces data at 100mbps but the
receiver can process data at only I Mbps, the transmission will overload receiver
and some data will be lost.

Heterogeneity in Distributed Systems

Heterogeneity describes the difference in network standards, operating system,
software, hardware and programming languages. These differences can be overcome by
developing some standards that regulate the interpretability of these differences. One
major technique used to overcome this problem is the divide and conquer rule. This
divides the system into three horizontal layers of operating system, middleware and
application. The middle layer hides the differences in environment to both upper and
lower layer. Examples of this layer are: Remote method invocation (RMI), common
object request broker architecture (CORBA), distributed component objects model
(DCOM), Network flight system (NFS), world-wide-web (WWW) and remote procedure
calls (RPC). The middle ware are set of software and hardware that work together at the
background to coordinate and integrate all the network, operating system and
programming language together.

Middleware is a layer of software between applications and OS that gives a uniform

interface. It is central to developing distributed applications.

32
 CSC413-Distributed Computing System

The role of middleware in distributed Computing System includes amongst others:

- It provides a layer of software offering a single-system view
-It offers transparencies (access, location, and so on)
-It simplifies development of distributed applications and services.

Assignment:

A. Discuss the operation of the following middleware layer as related to Distributed

Computing System.

1. Remote Method Invocation (RMI)

2. Distributed Component Object Model (DCOM)
3. Remote Procedure Call (RPC)
4. Common Object Request Broker Architecture (CORBA)

B. Discuss the Architecture of distributed systems

C. Discuss the components of a distributed

CLIENT-SERVER ARCHITECTURE
Client-Server Architecture: An information-passing scheme that works as follows: a client
program, such as Mosaic, sends a request to a server. The server takes the request,
disconnects from the client and processes the request. When the request is processed, the
server reconnects to the client program and the information is transferred to the client.
This architecture differs from traditional Internet databases where the client connects to
the server and runs the program from the remote site.

CLIENT SERVER NETWORK MODEL

In the client/server network model a computer plays a centralized role and is known as a
server. All other computers in the network are known as clients. All client computers
access the server simultaneously for files, database, docs, spreadsheets, web pages and
resources like hard diver, printer, fax modem, CD/DVD ROM and others.
In other words, all the client computes depends on the server and if the server fails to
respond or crash then networking/communication between the server and the client
computer stops.

33
 CSC413-Distributed Computing System

Client/server model is a concept for describing communications between computing

processes that are classified as service consumers (clients) and service providers (servers).
Figure 8 presents a simple C/S model.

The basic features of a C/S model are:

1. Clients and servers are functional modules with well-defined interfaces (i.e., they hide
internal information). The functions performed by a client and a server can be
implemented by a set of software modules, hardware components, or a combination
thereof.
2. Each client/server relationship is established between two functional modules when
one module (client) initiates a service request and the other (server) chooses to respond
to the service request.
For a given service request (SR), clients and servers do not reverse roles (i.e., a client stays
a client and a server stays a server).
However, a server for SRR1 may become a client for SRR2 when it issues requests to another
server. For example, a client may issue an SR that may generate other SRs.

3. Information exchange between clients and servers is strictly through messages (i.e., no
information is exchanged through global variables). The service request and additional
information is placed into a message that is sent to the server.
The server's response is similarly another message that is sent back to the client. This is
an extremely crucial feature of C/S model.

4. Messages exchanged are typically interactive. In other words C/S model does not
support an off-line process. There are a few exceptions. For example, message queuing
systems allow clients to store messages on a queue to be picked up asynchronously by the
servers at a later stage.

5. Clients and servers typically reside on separate machines connected through a network.
Conceptually, clients and servers may run on the same machine or on separate machines.

34
 CSC413-Distributed Computing System

The implication of the last two features is that C/S service requests are real-time messages
that are exchanged through network services. This feature increases the appeal of the C/S
model (i.e., flexibility, scalability) but introduces several technical issues such as
portability, interoperability, security, and performance.

Figure 8: Conceptual Client/Server Model

Characteristics of a Client
 Request sender is known as client
 Initiates requests
 Waits for and receives replies.
 Usually connects to a small number of servers at one time
 Typically interacts directly with end-users using a graphical user interface

Characteristics of a Server
 Receiver of request which is send by client is known as server
 Passive (slave)
 Waits for requests from clients
 Upon receipt of requests, processes them and then serves replies
 Usually accepts connections from a large number of clients
 Typically does not interact directly with end-users

Note

35
 CSC413-Distributed Computing System

Another type of network architecture is known as peer-to-peer,, in which each node or

instance of the program can simultaneously act as both a client and a server, and because
each has equivalent responsibilities and status.
Peer-to-peer architectures are often abbreviated using the acronym P2P. Both client-
server and P2P architectures are in wide usage today.
The Client Model
In the client-server model, processes are categorized as servers and clients. Servers are
the processes that provide services. Processes that need services are referred to as clients.

A client process needing a service (e.g. reading data from a file) sends a message to the
server and waits for a reply message. The server process, after performing the requested
task, sends the result in the form of a reply message to the client process.
Note that servers merely respond to the request of the clients, and do not typically initiate
conversations with clients.
In systems with multiple serves, it is desirable that when providing services to clients, the
locations and conversations among the servers are transparent to the clients.
Clients typically make use of a cache to minimize the frequency of sending data requests
to the servers.

Figure 9: Server manages a shareable resource used by a number of clients in the

network. The users log on to the client machines.

36
 CSC413-Distributed Computing System

Server Model
The server software architecture interact as superior and not subordinate. The client
process always initiates requests, and the server process always responds to request.
Theoretically, client and server can reside on the same machine.
Typically, however, they run on separate machines linked by a local area network. Many
different types of server exists, though most people tend to equate the term client-server
with a network of PCs or workstations connected by a network to a remote machine
running a database server. Examples of server are:
1. Mail servers.
2. Print servers.
3. File Servers.
4. Database (SQL) Servers

IMPLEMENTATION OF CLIENT – SEVER MODEL

The client/server model is a form of distributed computing where one program (the
client) communicates with another program (the server) for the purpose of exchanging
information.

The client's responsibility is usually to:

1. Handle the user interface.
2. Translate the user's request into the desired protocol.
3. Send the request to the server.
4. Wait for the server's response.
5. Translate the response into "human-readable" results.
6. Present the results to the user.

The server's functions include:

1. Listen for a client's query.
2. Process that query.
3. Return the results back to the client.

37
 CSC413-Distributed Computing System

A typical client/server interaction goes like this:

1. The user runs client software to create a query.
2. The client connects to the server.
3. The client sends the query to the server.
4. The server analyzes the query.
5. The server computes the results of the query.
6. The server sends the results to the client.
7. The client presents the results to the user.
8. Repeat as necessary.

Figure 10: A typical client/server interaction

Flexible user interface development is the most obvious advantage of client/server

computing.
It is possible to create an interface that is independent of the server hosting the data.
Therefore, the user interface of a client/server application can be written on a Macintosh
and the server can be written on a mainframe. Clients could also be written for DOS- or
UNIX-based computers. This allows information to be stored in a central server and
disseminated to different types of remote computers.
Since the user interface is the responsibility of the client, the server has more computing
resources to spend on analyzing queries and disseminating information.

This is another major advantage of client/server computing; it tends to use the strengths
of divergent computing platforms to create more powerful applications.

38
 CSC413-Distributed Computing System

In a nutshell, client/server computing provides a mechanism for distinct computers to

cooperate on a single computing task.

TIERED ARCHITECTURE
Generic client/server architecture has two types of nodes on the network: clients and
servers. As a result, these generic architectures are sometimes referred to as "two-tier"
architectures.

Some networks will consist of three different kinds of nodes: server, application servers
which process data for the clients and database servers which store data for the
application servers. This is called three-tier architecture.

The advantage of an n-tier architecture compared with a two-tier architecture (or a three-
tier with a two-tier) is that it separates out the processing that occurs to better balance
the load on the different servers; it is more scalable.

The disadvantages of n-tier architectures are:

1. It puts a more load on the network.
2. It is much more difficult to program and test software than in two-tier architecture
because more devices have to communicate to complete a user’s transaction.

First generation systems are 2-tiered architectures where a client presents a graphical
user interface (GUI) to the user, and acts according to the user's actions to perform
requests of a database server running on a different machine.

2-Tier Architectures.
Client/server applications started with a simple, 2-tiered model consisting of a client and
an application server. The most common implementation is a 'fat' client - 'thin' server
architecture, placing application logic in the client. As seen in Figure 11, the database
simply reports the results of queries implemented via dynamic SQL using a call level
interface (CLI) such as Microsoft's Open Database Connectivity (ODBC).
39
 CSC413-Distributed Computing System

Figure 11: Traditional FAT client/server development

Alternate approach is to use thin client - fat server waylays that invokes procedures stored
at the database server as shown in Figure 12. The term thin client generally refers to user
devices whose functionality is minimized, either to reduce the cost of ownership per
desktop or to provide more user flexibility and mobility.

In either case, presentation is handled exclusively by the client, processing is split

between client and server, and data is stored on and accessed through the server.

Remote database transport protocols such as SQL-Net are used to carry the transaction.
The network 'footprint' is very large per query so that the effective bandwidth of the
network, and thus the corresponding number of users who can effectively use the
network, is reduced.
Furthermore, network transaction size and query transaction speed is slowed by this
heavy interaction.
These architectures are not intended for mission critical applications

Figure 12: Thin client/server development

40
 CSC413-Distributed Computing System

Development tools that generate 2-tiered fat client implementations include:

PowerBuilder, Delphi, Visual Basic, and Uniface.
The fat server approach, using stored procedures is more effective in gaining
performance, because the network footprint, although still heavy, is lighter than that of a
fat client.

Advantages of 2-Tier System

1. Good application development speed
2. Most tools for 2-tier are very robust
3. Two-tier architectures work well in relatively homogeneous environments with fairly
static business rules
A new generation of client/server implementation takes this a step further and adds a
middle tier to achieve a '3-tier' architecture. Generally, client-server can be implemented
in an 'N-tier' architecture where application logic is partitioned. This leads to faster
network communications, greater reliability, and greater overall performance.

3-Tier Architectures
Inserting a middle tier in between a client and server achieves a 3-tier configuration. The
Components of three-tiered architecture are divided into three layers:
A presentation layer, functionality layer, and data layer, which must be logically separate.
As seen in Figure 13, the 3-tier architecture attempts to overcome some of the limitations
of 2-tier schemes by separating presentation, processing, and data into separate distinct
entities.
The middle-tier servers are typically coded in a highly portable, non-proprietary language
such as C.
Middle-tier functionality servers may be multithreaded and can be accessed by multiple
clients, even those from separate applications.

41
 CSC413-Distributed Computing System

Figure 13: 3 - Tiered Application Architecture

The client interacts with the middle tier via a standard protocol such as dynamic link
library (DLL), API, or RPC. The middle-tier interacts with the server via standard
database protocols.
The middle-tier contains most of the application logic, translating client calls into
database queries and other actions, and translating data from the database into client data
in return.

If the middle tier is located on the same host as the database, it can be tightly bound to
the database via an embedded third generation language (3gl) interface. This yields a very
highly controlled and high performance interaction, thus avoiding the costly processing
and network overhead of SQL-Net, ODBC, or other CLIs.

Furthermore, the middle tier can be distributed to a third host to gain processing power
capability.

Advantages of 3-Tier Architecture

1. RPC calls provide greater overall system flexibility than SQL calls in 2-tier architectures
2. 3-tier presentation client is not required to understand SQL. This allows firms to access
legacy data, and simplifies the introduction of new data base technologies
3. Provides for more flexible resource allocation. Modularly designed middle-tier code
modules can be reused by several applications

42
 CSC413-Distributed Computing System

4. 3-tier systems such as Open Software Foundation's Distributed Computing

Environment (OSF/DCE) offers additional features to support distributed applications
development

As more users access applications remotely for business-critical functions, the ability of
servers to scale becomes the key determinant of end-to-end performance. There are
several ways to address this ever-increasing load on servers. Three techniques are widely
used:
 Upsizing the servers
 Deploying clustered servers
 Partitioning server functions into a "tiered" arrangement

N-Tier Architectures
The 3-tier architecture can be extended to N-tiers when the middle tier provides
connections to various types of services, integrating and coupling them to the client, and
to each other. Partitioning the application logic among various hosts can also create an
N-tiered system.
Encapsulation of distributed functionality in such a manner provides significant
advantages such as reusability, and thus reliability.

As applications become Web-oriented, Web server front ends can be used to offload the
networking required to service user requests, providing more scalability and introducing
points of functional optimization. In this architecture, as seen in Figure 14, the client
sends HTTP requests for content and presents the responses provided by the application
system.

On receiving requests, the Web server either returns the content directly or passes it on
to a specific application server. The application server might then run CGI scripts for
dynamic content, parse database requests, or assemble formatted responses to client
queries, accessing dates or files as needed from a back-end database server or a file server.

43
 CSC413-Distributed Computing System

Figure 14: Web Oriented N-Tiered Architecture

By segregating each function, system bottlenecks can be more easily identified and
cleared by scaling the particular layer that is causing the bottleneck.

For example, if the Web server layer is the bottleneck, multiple Web servers can be
deployed, with an appropriate server load-balancing solution to ensure effective load
balancing across the servers as illustrated in Figure 15.

Figure 15:-Tiered Architecture with server load balancing

The N-tiered approach has several benefits:

1. Different aspects of the application can be developed and rolled out independently.
2. Servers can be optimized separately for database and application server functions
3. Servers can be sized appropriately for the requirements of each tier of the architecture
4. More overall server horsepower can be deployed

44
 CSC413-Distributed Computing System

5. They are far more scalable, since they balance and distribute the processing load among
multiple, often redundant, specialized server nodes.
6. It turn improves overall system performance and reliability, since more of the
processing load can be accommodated simultaneously.

The disadvantages of n-tiered architectures include:

1. More load on the network itself, due to a greater amount of network traffic.
2. More difficult to program and test than in two-tier architectures because

45