Welcome to Lesson 10, Day 2- Chapter 9 - Understanding Complex Networks
Centralized
vs. Client/Server Computing
Client/server computing evolved from
the centralized computing environment that was first introduced in the 1950s
and 60s and was so common on up into the 1970s and 80s. During this time most
computing was done on large mainframes via "dumb" terminals connected
directly to the mainframe. Mainframe computing still goes on today with PCs
and "thin clients" attaching to terminal servers of some kind in order
to access the mainframe. The primary drawback to this type of computing is the
slowdown often experienced by the traffic created on the network by all the
processes traveling back and forth between the PC/thin client and the mainframe
(for every keystroke a packet is sent across the network and then a response
is sent back.)
Understanding
Terminal Services
Terminal Services allows clients to run large or complex applications on older,
less-capable clients or thin clients. To do this, the clients utilize the processing
power of the server, rather than local processing. For this reason, Terminal
Services servers must be heavily configured with large amounts of RAM and disk
space, as well as multiple, very powerful CPUs.
Back to the Future:
The Mainframe Environment
M any of today’s client/server applications and environments operate in
the same manner as mainframe computers. This tried and true processing method
allows for efficient use of older hardware and, has survived throughout the
PC revolution.
We are seeing a return to this computing
model with the advent of application servers, centralized database and webs
servers and the like. Certain large-scale, transaction-intensive applications
are much more suited to this format.
Client/Server Environment
With client/server networks, some processing occurs on the server, and some
on the client. A good example of this is an application server which runs a
groupware product. The server manages the clients’ requests for files
and controls the groupware processing.
A server which performs only shared file services is often not considered a
client/server configuration. This is called a shared-file storage system.
Client/Server Model in a Database Environment
A DBMS environment is the penultimate example of client/server processing. The
client makes a request for data to the server. The server searches through the
database to get the information requested and returns it to the client. The
client, then, is able to perform calculations, sorting, and filtering on the
data.
I n all client/server environments, the front-end is the client, and the back-end
is the server. The client uses SQL to translate the user's request into a format
the database can understand. SQL (structured query language) was developed by
IBM to provide simple, English-based commands that are issued to manipulate
data instead of having to use the cryptic programming languages that had been
used prior to SQL's introduction. SQL has become the standard in database queries
in general. Oracle is one
of the premier database management systems whose base language is SQL.
Client/Server Architecture
The single database server and distributed database architectures are pictured
in your text on p. 336, Figure 9-6. When establishing these types of networking
environments, the client musbe properly configured to support the server structure.
When distributed databases are installed, the servers often act as a single
entity, so that client configuration may actually be the same as with a single-server
network.
Advantages of Working in a Client/Server Environment
The primary reason is distributed processing. This utilizes the superior computing
power of the server for the most arduous tasks, but also uses the CPU on the
client to perform minor calculations.
Click
Here for the MiniLecture on Application Service Providers
Remember to keep up with your on-line lessons and quizzes!
mdm