InfoCenter Home >
7: Multimachine management >
7.1: Using WebSphere Application Server in a multimachine environment >
7.1.1: Scaling up WebSphere applications
7.1.1: Scaling up WebSphere applications
Multimachine applications can be configured in a variety of ways to scale up a system
to add more processing power, improve security, maximize availability, and balance
workloads. The WebSphere Application Server, Advanced Edition provides several ways to implement
configurations that address these issues. These scaling techniques are generally combined
to maximize the benefits and minimize the problems associated with multimachine systems.
- Cloning. Cloning allows the creation of multiple copies of an object
such as a servlet, enterprise bean, or entire application server. The first step is to
create a model of the object based upon the object's current configuration. From the
model, you can then create clones of that object. Clones can be created on the same
physical machineor on different machines. Using clones can improve the performance of a
server, simplify its administration, and enable the use of workload managment; however,
there is a point of diminishing returns when adding more clones slows down the system due
to the extra network traffic required for managing the clones.
- Workload management (WLM). Incoming processing requests from clients
are transparently distributed among the clones of an application server. WLM enables both
load balancing and failover, improving the reliability and scalability of WebSphere
applications. In addition, administrative servers can participate in WLM for failover
support.
- Open Servlet Engine (OSE). A Web server is responsible for receiving
client requests, filtering them, and forwarding them to the servlet engine in an
application server for processing. This forwarding is accomplished through a transport
mechanism called OSE. OSE is a proprietary internal protocol that uses IPC mechanisms
provided by the underlying operating system to transport data. OSE can be used to perform
two types of load distribution in WebSphere. It can be configured to forward different
URLs to different application servers, and it automatically distributes client requests
among all available clones of a servlet (whether local, remote, or both). OSE
automatically handles failover and changes in the available clones.
- Servlet redirection. A servlet redirector is a special-purpose
application server that forwards HTTP requests from clients to a servlet engine in a
cloned application server. It can participate in workload management for load balancing
and failover support.
- IP sprayer. An IP sprayer transparently redirects incoming HTTP
requests from Web clients to a set of Web servers. Although the clients behave as if they
are communicating directly with a given Web server, the IP sprayer is actually
intercepting all requests and distributing them among all the available Web servers in the
cluster. IP sprayers (such as IBM Network Dispatcher or Cisco Local Director) can provide
scalability, load balancing, and failover for Web servers.
|
|