By jeckels on Oct 30, 2014
We held a webcast last week (now available free on-demand) on the most common use cases for in-memory data grids, and it was one of the year's most-attended online events. You all kept us busy answering your questions and introducing you to concepts that can help you solve some of your core IT challenges.
We thought it'd be useful to record some of the questions and answers we received just in case you missed them (some questions lightly edited for clarity).
|How big is the Coherence footprint ? Does it support Virtual and physical hardware?||The software library itself is pretty small, most of the footprint is for storing data. It does support both physical and virtual.|
|Backend: which languages does it support? example: java , pojo, .net...||Java, .NET, C , REST, memcached clients, JCache Clients (JSR-107)|
|Will cache updating work with version 11gR2?||I think I understand your question. You want to know how changes to the database are reflected in the cache. In Coherence 12c, we introduced GoldenGate HotCache, which listens to database change logs and pushes changes into Coherence. This will work with version 11gR2.|
|Once data is cached on Coherence, what is the criteria to offload the data, will it be taken care of by Coherence, or we need to provide any specific logic/criteria?||Offload works by accessing data in a key-based manner. If your application is SQL based and you don't want to change it, Coherence may not be a good fit you, unless you are using some persistence technology like JPA.|
|What is a Coherence application? This was mentioned in the initial part of the presentation.||By Coherence application, we mean an application that uses Coherence. Probably a bad term - some large apps are built on Coherence as if Coherence itself were the app server!|
|From what I understood, Coherence works with multiple application servers like Oracle WebLogic. Is there any advantage to using Coherence with Oracle WebLogic as compared to JBoss/Tomcat or any other application servers?||Yes, we have built-in integration for areas like lifecycle management and monitoring, application deployment, security, simpler session management, and more coming in future releases.|
|If Cohernece is down due to some reason, will the application will be able to access the original data?||If you use Coherence as a "look aside" cache, yes. But most customers do not do this and rely on Coherence being available, together with a reasonable DR and "Maximum Availability Architecture" strategy.|
|How much data can be cached by Coherence without effecting its performance?||Coherence is designed to be horizontally scalable. Customers run 100s of cluster members in some cases, and even smaller clusters can cache TBs of data with no performance impact.|
|Do we have any reporting machanisim on the caching to get details about highly accessed data & how often?||OOTB, Not on a per object basics, but at a certain higher level. We are adding auditing functionality to Coherence soon that may enable you to track that. Today, though, we have hooks where you can insert your own logic to do this.|
You can see the webcast in its entirety on-demand, or if you have further questions please feel free to comment here.