Clever Geek Handbook

Cassandra as a distributed cache storage

Can Cassandra be used as a distributed in-memory cache database using its file-level caching, cache cache, and line cache?

I don't want to overload every node, and I want to add more nodes to the cluster as the data grows to make it efficient (so that most of my data is cached). Especially since 40% of my column families are static and there aren't many updates / inserts to other tables.

Our main task is that we need to take place in real time in real time (faster than in dB memory)

+3


source to share


3 answers


Cassandra was not born for purpose, but after many optimizations, she became an in-memory caching tool. There are a few experiments - the most important one I know is the post reported by Netflix. At Netflix, they replaced their system with a new cache architecture based on SSD Cassandra - the results are very impressive in terms of performance improvements and cost savings.

Before choosing Cassandra as a replacement for any caching system, I would recommend a deep understanding of the use of string caching and key caching. Moreover, I have never used Datastax Enterprise, but it has an interesting memory table .



NTN, Carlo

+6


source


I think you could, but I don't think that's the correct use case for Cassandra. Without knowing any more about your requirements, I would recommend that you take a look at products like for example. Hazelcast , which is a distributed in-memory cache and sounds more like a good fit for your use case.



+3


source


I know it a little late, but I just came to this post doing some research in Cassandra.

I have seen success with Tibco AST (recently renamed DTM) for memory caching.

I also played around with the Pivotal gemfire (this uses Geode under the covers) which showed some promise.

+1


source






More articles:


All Articles