Posted by arlene

Database Servers

Major database vendors offer software features that work within networks to provide one or more database servers. Before database servers were available, all database access on a local ma‑chine required that the database reside physically on the same machine. It was very difficult, if not impossible, to access data on any machine other than where the data resided.

Database software was later designed to use file servers to store the data, but this was very costly in loss of performance and resulted in extremely high network traffic. The availability of database servers has made central and distributed storage a practical reality. Database servers and the advertising of database vendors has recently made database servers and client/server technology synonymous to many.

Database servers make database functions available on a server. File access by the database software is performed locally on the server. With database servers, the application runs on the work station, accessing the database with requests that execute database services in the server. A single database server can support a number of work stations.

Some databases provide stored procedures, a feature that has a limited capability to perform application functions within the database. Stored procedures provide a pseudo-distributed processing capability that can be quite powerful.

Database software running on the client system intercepts requests for database access and uses the network software to route the requests to the database software running on the server. Database software executes the request on the server and returns the result to the client. Since these types of systems require close cooperation between the network software and the database software, database vendors must write special software for each network system supported.

Enhanced Client/Server Processing

Enhanced C/S processing adds the following features:

• Application servers may be placed on any system in the network.
• The load is automatically balanced among several servers performing the same service.
• Multiple application servers may be active on the same or multiple machines.
• Requests may be routed to a particular server, depending on the data passed with the request.
• Application programs are never aware of the locations of services invoked.
• Multiple database types (heterogeneous databases) can be
  used in the system with a minimal programming effort.

• New types of databases may be added, or existing database types may be changed with little or no impact on application programs.

Enhanced client/server processing is provided by distributed transaction processing systems. It is unfortunate that many people think of transaction processing as a specific requirement: in fact, all programs that update databases are performing transaction processing.

One or more clients issue requests for service that are intercepted by the name server. The name server routes the request to the appropriate server based on the name of the service and other considerations. The server performs the requested service and returns the response to the client. Multiple servers may provide the same service. The name server provides several methods to determine which server to use including data-dependent routing and load balancing. Both methods may be changed by administrative services without affecting the operation of the clients or servers.

Possibly related posts: (automatically generated)
Network and Servers Technical Compare continue…

• Network and Servers Technical Compare
• Dynamic DNS
• Client/Server must Know
• Directory Replication
• Primary, Secondary, and Caching-Only Name Servers
• What Role Will Your Server Perform?
• What Role Will Your Server Perform?
• Network Access Control Databases
• Website Hosting Sever, some Pitfalls you need to avoid part 1
• UNIX Servers Typical Hardware