Terracotta Discussion Forums (LEGACY READ-ONLY ARCHIVE)

I am trying to understand the functional differences (if any) of using BigMemory independently from Ehcache. The documentation for ehcache configuraiton is much more robust than that for BigMemory. Does that mean BigMemory alone is less capable than using BigMemory as a plug-in for Ehcache? Sorry for the newbie question but the answers are not clear to me from the documentation. Understanding just what I can do for transaction control, eviction, etc.. in BigMemory is important to me. Thanks for any guidance.

BigMemory always runs with Ehcache as its get/put interface. So when you're running BigMemory, you're also running Ehcache.

With Ehacache alone, you're going to see garbage collection pauses start degrading performance once you've got a few GB in your JVM's heap. When you move up to BigMemory, you can scale to whatever RAM you've got on the machine (BigMemory Go) -- or to whatever RAM you can put in a server array (with BigMemory Max). BigMemory handles the data consistency, persistent backup, and all the rest. Plus, with BigMemory you're getting the Terracotta Management Console with advanced monitoring of in-memory stores, fast restartability, and a lot more.

So, the decision to use BigMemory is first and foremost about the volume of data you want to keep in memory without degrading performance. If you're talking over a few GB -- and/or if you want a distributed in-memory store -- BigMemory is the way to go. You also might want to take advantage of BigMemory's more robust in-memory management features.

We recently did a webcast about what you get with BigMemory Go over Ehcache -- you can check out the recording here:
http://www.terracotta.org/resources/webcasts/5-reasons-to-upgrade-ehcache-to-bigmemory-go

To check out all the features in BigMemory, check out http://terracotta.org/bigmemory

Thanks for the information and I have seen the webinar mentioned. That is one thing that made me want to sort out the differences between Ehcache and BigMemory

I am looking at use cases where there is more need to store data than can be accomplished using heap only. So I get the usage of BigMemory as a data store. I am also seeing that using BigMemory Go or Max for clustering allows me to get large stores up and running. However, when it comes to setting things up like transactions, write-through or write-behind configuration, and all the other kinds of configuration possible with Enterprise Ehcache it still isn't 100% clear to me if I can do all that using BigMemory alone. There is always lots to learn with a new piece of technology and it can take a while to figure it all out. Figuring out what you can or can't do with various technology mixes makes it just a little bit harder.

Hi,
Sorry for the confusion. The features you described are configured and work the same whether using BigMemory or Ehcache.

Does that help (Making sure I understand the question)?

I believe that does help. Sounds like BigMemory does all that Ehcache does plus adds in off-heap memory storage. Along that line of logic, when I read that BigMemory can be used as a plug-in store for Ehcache I can take that to mean an option targeting existing Ehcache users. If I am new to licensing or subscribing (whole different issue) Terracotta products then BigMemory Max is all I need to license for my clustered caching needs.

Pretty much. Also adds fault tolerance and an admin console.

Yeah, licensing big memory max will take care of all your clustering needs.

Sorry for the confusion