CirrusI am trying to understand what is the key difference between the Web Acceleration (Cache) and Ram Cache on the BIG-IP. I am reading the concepts of both, and they are pretty much the same. One of the key differences I noticed, is that Web Acceleration (Cache) stores HTTP GET responses only and we can set it to store URIs, while the Ram Cache can cache HTTP objects. My main question I think is: When to use one or the other?
Cirrocumulus
CirrusThat's what I am trying to understand. The 301a Blueprint mention both features as they are independent; however, the definition is the same for both according to my understand. I just want to check if I am missing something regarding these features.
CirrocumulusCirrusI just got an answer from one of my company's F5 instructors: In fact, it's very simple. LTM Ram Cache = uses the unallocated RAM space on your BIG-IP. Each time you add the Web Acceleration profile, it takes the RAM you assigned from the unallocated space of RAM. So at some point, you can run out. It doesn't regroup it and doesn't use your disk. WA aka Web Acceleration aka AAM uses ONLY it's reserved RAM portion that you reserve when you provision it. When there isn't enough RAM it sends it down to the disk space that it reserves when you provisioned that module. So all in all, if you want to do a very simple RAM Cache on the BIG-IP for one or a few Virtual Servers, then LTM RAM Cache is enough. If you want to have real Caching on the BIG-IP that uses both RAM and Disk and can serve a bunch of Virtual Servers, then purchase AAM.
Cirrocumulus