OracleAS Portal meets the three critical requirements for scaling
performance:
The underlying architecture is cross-platform, allowing the portal to take
advantage of the full range of available hardware resources. Portal
administrators can choose the best platform for each portal component,
allowing optimization of both performance and total cost of ownership.
The architecture supports load distribution and parallel execution of
portal components across multiple servers.
The portal implements intelligent caching of dynamically generated portal
pages. Intelligent caching ensures that information remains fresh and timely
while minimizing the cost of regenerating content from databases and back-end
services. Intelligent caching supports user-level customizations and is also
modular, so that entire pages or just page components can be stored in a cache
and refreshed as necessary. Finally, intelligent caching protects security by
ensuring that cached content is only accessible by authorized users.
At the heart of the OracleAS Portal architecture is the Parallel Page Engine,
a multithreaded Servlet deployed on OracleAS Containers for J2EE (OC4J),
Oracle's highly scalable J2EE framework. Multiple Parallel Page Engines deployed
in a server farm work together to retrieve content from portlet providers,
manage caching, and assemble and deliver pages.
OracleAS Portal is also fully integrated with OracleAS Web Cache, Oracle's
patented caching technology. Unlike legacy cache servers, which only handle
static data, OracleAS Web Cache combines caching, compression and assembly
technologies to accelerate the delivery of both static and dynamically generated
Portal content. OracleAS Web Cache also provides back-end Web server load
balancing, failover and surge protection features which ensure blazing
performance and rock-solid up-time. With OracleAS Web Cache, OracleAS Portal can
now serve rich content faster, to more users, using fewer computing resources
than ever before.
Without the unique combination of caching and scalability features offered by
OracleAS Portal, it would be impossible to scale a portal site in a
cost-effective manner. OracleAS Portal's architecture truly minimizes the
hardware resources required to deploy a high-traffic portal.
Measuring the scalability of a portal requires for there to be common
denominator, in the case of Database performance metrics we can refer to TPCC
benchmarks. For J2EE application server performance we can refer to 'Pet Store'
transaction figures. Unfortunately there is no 'Pet Store' for Portals, until
the true unification of portal development and deployment standards through the
efforts of JSR168, WSRP and other open portal development standards, it will be
impossible to develop a Portal Pet Store because of the variety of
implementation methods employed by the portal vendors in the marketplace.
This document provides links to documents related directly to
questions you may have about the sizing, performance and scalability of OracleAS Portal.
Common Sizing, Performance and Monitoring Questions
How do I estimate the hardware I will need to run the portal ?
You can read Sizing
OracleAS Portal - An Introduction, which will explain the things
that need to be considered when estimating the hardware you'll need to deploy a
portal solution.
Having read the sizing overview, you can see if your portal requirements fit
these sample architectures
Do you have any examples of existing portal implementations ?
Currently we have two detailed examples that you can read about. We are
working on delivering more real world examples that have hardware and
performance specifications
My portal is cool ! I'd like to tell Oracle all about it...
Great, we'd love to hear about it.
What other forms of monitoring are available for my portal ?
Administer
and Monitor Your Portal with Oracle Enterprise Manager (40049) -
this whitepaper describes how a portal administrator can take advantage of
Oracle Enterprise Manager features such as collections, automatic notifications,
historical metric reports, application modeling and application performance
monitoring. We will show how to configure and interpret diagnostic logging
information by using the powerful new Oracle Enterprise Manager Log Viewer for
automatic correlation of these logs.
You can also read the following technotes for further monitoring advice:
How do I know what is being invalidated in Web Cache ?
Good question, you can find out by reading Managing
Cache Invalidation with OracleAS Portal This is important if you
are seeing slow performance from invalidations or perhaps stale content that
you'd like to invalidate manually
I've heard about this thing called TCP.NODELAY - what is it ?
What sort of settings should I change to tune the portal ?
Ideally, the portal installation should provide you with reasonable
performance out of the box, but if you feel the need to adjust elements of the
OracleAS Portal installation then the following papers will assist you
Tips
and Techniques for Deploying a Scalable Portal (40048) -
this whitepaper describes the unique OracleAS Portal features that enable
scalable performance, focusing on the process used to generate, assemble,
secure, and cache personalized portal pages.
I've got some questions about sizing and performance - who do I ask ?
You should ask them on the forums, specifically the
Performance Forum
or if it's related to caching performance and or strategies, you can ask it on
the Web Cache Forum
Bookmark
