Other Caching Mechanisms

Web server caching

Most Web servers can cache static files. IBM HTTP Server, which comes with WebSphere Application Server, has a facility called the Fast Response Cache Accelerator (FRCA) on AIX and Adaptive Fast-Path Architecture (AFPA) on Windows that can cache static content (both platforms) as well as dynamic content (Windows only). This is a kernel-based cache and, by far, the fastest cache of any discussed here.


Edge Side Include (Web server plugin caching)

The ESI processor can cache pages or fragments of pages at the HTTP server layer.

Each time a new request is received by the plug-in, the ESI processor checks for it in the cache. If some fragments are already in cache, the plug-in can use it. If not, the ESI processor adds a specific header named Surrogate-Capabilities before forwarding the request to the application server.

The application server responds to the request. If servlet caching is enabled in the application server and the output is edge cacheable, the application server adds a Surrogate-Capabilities header with caching information. The plug-in stores, in the cache, the application responses, builds the page with all the nested components, and returns the answer to the client.


Hardware caching

Multiple network equipment providers offer hardware cache devices. These devices serve the same purpose as software caches do, namely to offload content. The main difference is that these appliances are not running full versions of an operating system. Instead, they use a specialized operating system that is dedicated to performing the caching function.

This operating system can include custom file systems (that offer higher performance than the operating system file system) and a significantly reduced instruction set. By placing dedicated appliances instead of software caching in your architecture, you can reduce total cost of ownership, because these appliances do not have to be managed as strictly as machines with full operating systems.