Page Three

By Andrew Garcia  |  Posted 2004-11-22 Print this article Print

We liked Big-IP Version 9s use of nested profiles to ease ongoing administration. Administrators define profiles that dictate configuration settings on a virtual server for various services or protocols, in addition to encryption, authentication or TCP offload parameters. Nested profiles allowed us to easily set basic configuration levels for our virtual servers, which cascade to any tailored child profiles we defined to address changes that a virtual server might demand.

Big-IP provides bidirectional SSL acceleration. Companies can choose to terminate encryption from client machines at the Big-IP to allow application processing and reduce load on the servers, and they can re-encrypt data between the Big-IP and Web servers to provide maximum data security.

F5s OneConnect TCP offload functionality allows Big-IP to establish and maintain TCP sessions with the back-end servers and multiplex multiple client connections within these sessions. This reduces the amount of TCP processing overhead on the servers, which should increase the number of simultaneous client connections each server can support.

During tests, we noticed that Big-IP does not currently support caching of Web content. F5 officials said that simple caching poses too much of a risk for serving stale content and that advanced routines are necessary to avoid this. Advanced caching features in Big-IP should be available next year, according to the officials.

We appreciated the depth of control we could get over compression settings. By default, Big-IP allows administrators to select compression depending on URL extension or data type, and iRules allows even more flexibility. The compression level can be adjusted, depending on how much processor power administrators want to dedicate to the feature.

Compression levels varied according to the content in our tests of three page types (static HTML only, a mix of HTML and GIFs, and a dynamically generated Web page) at two compression settings. However, we generally found that the extra processing power needed for the highest levels of compression yielded minimal improvement.

The static page compressed 64 percent at the lowest setting and 69 percent at the highest. The mixed page achieved 41 percent compression at the low setting and 44 percent on high. The dynamically generated page showed 17.5 percent compression at both settings.

F5s user support site is impressive and includes an active user community where F5 customers discuss various issues and can post iRules configuration files to perform specific tasks—although there were a limited number of iRules available online during our testing.

Technical Analyst Andrew Garcia can be reached at

Check out eWEEK.coms for the latest news, views and analysis on servers, switches and networking protocols for the enterprise and small businesses.

Andrew cut his teeth as a systems administrator at the University of California, learning the ins and outs of server migration, Windows desktop management, Unix and Novell administration. After a tour of duty as a team leader for PC Magazine's Labs, Andrew turned to system integration - providing network, server, and desktop consulting services for small businesses throughout the Bay Area. With eWEEK Labs since 2003, Andrew concentrates on wireless networking technologies while moonlighting with Microsoft Windows, mobile devices and management, and unified communications. He produces product reviews, technology analysis and opinion pieces for, eWEEK magazine, and the Labs' Release Notes blog. Follow Andrew on Twitter at andrewrgarcia, or reach him by email at

Submit a Comment

Loading Comments...
Manage your Newsletters: Login   Register My Newsletters

Rocket Fuel