Client / Server Computing
What is Client/Server?
Client/Server (referred to as C/S for brevity) is a logical progression from from modular
processing, where different program modules are responsible for different tasks. Client/server is
a computational architecture that involves client processes requesting service from server
processes. In the client/server model, these processes are generally maintained on different
pieces of hardware. But it is important to remember that the client and server are both
SOFTWARE processes!! Although many generally call hardware boxes 'servers' or 'clients', the
clients or servers are actually the software processes these computers are running.
Even thought the client and server processes are separated, a C/S application should seem like it
operates as a single application to the user.
The CLIENT: The client is a process (program) that sends a message to a server process
(program), requesting that the server perform a task (service). Client programs usually manage
the user-interface portion of the application, validate data entered by the user, dispatch requests
to server programs, and sometimes execute business logic. The client-based process is the user's
point of interaction with the application. The client process contains solution-specific logic and
provides the interface between the user and the rest of the application system. The client process
also manages the local resources that the user interacts with such as the monitor, keyboard,
workstation CPU and peripherals. One of the key elements of a client workstation is the
graphical user interface (GUI). Normally a part of operating system i.e. the window manager
detects user actions, manages the windows on the display and displays the data in the windows.
The SERVER: A server process (program) fulfills the client request by performing the task
requested. Server programs generally receive requests from client programs, execute database
retrieval and updates, manage data integrity and dispatch responses to client requests. Sometimes
server programs execute common or complex business logic. The server-based process "may"
run on another machine on the network. This server could be the host operating system or
network file server; the server is then provided both file system services and application services.
Or in some cases, another desktop machine provides the application services. The server process
acts as a software engine that manages shared resources such as databases, printers,
communication links, or high powered-processors. The server process performs the back-end
tasks that are common to similar applications.
Categories of Component Processes in the Client/Server Model
Three categories:

The User Interface (UI): how the application is presented to the user;

The Business Logic (BL): the processing rules that are used in the C/S application; a
payroll application will have different business rules than a trading floor application, for
example;

Data Management (DM): the storage of the data used by the application- which
implements the information model
These three components can be divided among the client and server processes in several ways-See the table below:
Types of Client/Server Computing
The Gartner Group came out with the five ways of describing the different c/s styles based on
how they split the three components of any application: user interface, business or application
logic, data management. The five styles are distributed presentation, remote presentation,
distributed function, remote data management, and distributed data management. (Note: This is
an arbitary classification and others may do it differently).
Types of Client/Server Computing
C/S Type
Distributed
Presentation
Remote
Presentation
Distributed
Logic
Data Mgt.
Data Mgt.
Data Mgt.
Functions on
Business Logic Business Logic Business Logic
Server Side
User Interface
Remote Data
Management
Distributed
Database
Data Mgt.
Data Mgt.
Network over which the processes communicate
Functions on
Client Side
Data Mgt.
Business Logic Business Logic Business Logic
User Interface User Interface User Interface User Interface User Interface
'Thinnest' Client ---------------------------------------------------------- 'Fattest' Client
Datamation, 4/1/95
Distributed and Remote Presentation -- For people whose roots are embedded in the
mainframe IBM world, client-server is essentially distributed or remote presentation. This style
maps a workstation Graphical User Interface (GUI) front end onto an existing application's textbased screen. In Distributed Presentation, we are generally looking at a system which uses a
dumb terminal at the client side. This is the extreme of the 'thin' client model (see below).
Distributed Logic -- Here the split occurs in the application functionality, one part going to the
client, other to the server. Distributed Logic applications are the most complex of the three basic
typologies since two separately compiled application programs must be developed. Developers
must analyze where each function should reside and what type of dialog must occur between the
two programs. The underlying communications facilities may implement either a message-based
or remote procedure call (RPC) mechanism for transfer of dialog and data.
Remote Data Management -- In remote data management, the entire application resides on the
client and the data managment is located on a remote server/host. Remote Data Management is
relatively easily to program for because there is just one application program.
Distributed Database -- This is when data management and application functions occur at both
the client and server. In this instance, data management at the client would include referential,
read-only data. Data frequently updated or accessed by many users would reside on the server.
This is the extreme of the 'fat' client model.
Fat Clients vs. Thin Clients
Fat Clients -- These are fully loaded PC's - desktops or laptops - containing a full suite of PC
operating system, Windows, PC applications and network connectivity software. This is a
misnomer, because clients are actually processes, not machines. These systems are ready to run
all types of processes--user interface, business logic, and data management. Machines that are
capable of running all these level of processes are costly and are complex to manage.
Thin Clients -- These are machines that will download what they need to run from a network.
They attach to a server and provide a graphical interface on a terminal that is optimized for
network-centric computing. Thin clients cost a fraction of the price of a fat client. Thin clients
can readily access the internet, network based application and other host based systems. These
devices are designed to take the complexity out of managing them. They will be managed like
dumb terminals--there will be no local data storage and no disk drives to worry about, yet users
will be able to select applications from various operating systems connected to the network
simultaneously.
Fat Client -- Pros (+) and Cons (-)

(+) More flexibility for the user;

(+) Since the client does more processing, the server does not have to be very powerful;

(-) Because the server has to deliver more data for the client to process, there are large
data sets going through the network, leading to greater network congestion;

(-) Machines capable of running fat client processes are more expensive, because they
require more processing power, more RAM, and more secondary storage;

(-) Fat clients increase managerial costs; changes to business logic have to be distributed
to all of the individual clients; there is much more user training involved; fat clients have
a variety of software that must be supported;

(-) The annual cost of ownership of each fat client has been estimated at $12K to $16K
per year; this includes all the real costs of the hardware and software, plus all the hidden
costs of software upgrades, managing the machines, and training the users for all the
different client processes they are using.
Thin Client -- Pros (+) and Cons (-)

(-) Less flexibility for the user;

(-) Because the server processes handle more of the processing, the machine running the
server processes has to be very powerful;

(+) The server only delivers the results of client queries, not entire data sets. Because of
these, the server is sending much less data through the network. Because of this, there is
generally less congestion on the network;

(+) Machines that run only thin client processes do not need as much processing power,
RAM, or secondary storage. Because of this, these thin client machines are less expensive

(+) Thin clients minimize some managerial costs; most changes can be made at the server
level, eliminating the need to distribute changes to each client machine.
Two-Tier and Three-Tier Architectures
Two-Tier Architecture The three components of an application (user interface, business logic,
and data management) are split into only tow tiers--the client and the server. In a two-tier
architecture, a client talks directly to a server, with no intervening server. It is typically used in
small environments (less than 50 users).
A common error in client/server development is to prototype an application in a small, two-tier
environment, and then scale up by simply adding more users to the server. This approach will
usually result in an ineffective system, as the server becomes overwhelmed. To properly scale to
hundreds or thousands of users, it is usually necessary to move to a three-tier architecture.
Three-Tier Architecture A three-tier architecture splits three parts into three tiers, and
introduces a server (or an "agent") between the client and the server. The role of the agent is
manyfold. It can provide translation services (as in adapting a legacy application on a mainframe
to a client/server environment), metering services (as in acting as a transaction monitor to limit
the number of simultaneous requests to a given server), or intellegent agent services (as in
mapping a request to a number of different servers, collating the results, and returning a single
response to the client. The goal of the three-tier architecture is to provide a middle-tier process,
called middleware, that focuses as an intermediary between the server and the client. This
allows the client process to emphasize the user interface, and the server side to emphasize data
management.
Type of Middleware
Database middleware is used on database-specific environments. It provides the link between
client and server when the client application that accesses data in the server's database is
designed to use only one database type.
Remote procedure calls (RPC) middleware is a more general-purpose solution to client/server
computing than database middleware. RPCs are used to access a wide variety of data resources
for use in a single application.
Messaging middleware takes the RPC philosophy one step further by addressing the problem of
failure in the client/server system. It provides synchronous or asynchronous connectivity
between client and server, so that messages can be either delivered instantly or stored and
forwarded as needed.
Object middleware delivers the benefits of object-oriented technology to distributed computing
in the form of object request brokers. ORBs package and manage distributed objects, which
can contain much more complex information about a distributed request than an RPC or most
messages and can be used specifically for unstructured or nonrelational data.
Transaction-processing (TP) monitors have evolved into a middleware technology that can
provide a single API for writing distributed applications. TP monitors generally come with a
robust set of management tools that add mainframelike controls to open distributed
environments.
Proprietary middleware is a part of many client/server development tools and large
client/server applications. It generally runs well with the specific tool or application environment
it is a part of, but it doesn't generally adapt well to existing client/server environments, tools, and
other applications.
Business Drivers for Client / Server Computing

Downsizing from 'big iron' (mainframe) environments;

Movement from vertical 'stovepipe' applications to enterprise-wide systems;

Organizations have changed from steep hierarchies to flattened hierarchies

Network management is replacing vertical management

There is a change to team based management

The user will perform as much processing as possible during customer contact time

Multi-skilled and multi-function teams need access to multiple applications
The development and implementation of client/server computing is more complex, more difficult
and more expensive than traditional, single process applications. The only answer to the question
"why build client/server applications?" is "because the business demands the increased benefits.
Send questions about this outline to Bill Ferns
Download an MS-Word 6.0 version of this file.
Return to the Bill Ferns' CIS9000 page.