Jan 182010
 

It seems Cassandra creates a big pile of threads for different tasks, I didn’t step into details, but I’m pretty sure it has more than 40 threads with default setting on a 2-nodes cluster. So multi core may not be a concern, as all these threads can run on different cores to fully utilize the CPU resource.

However, my tests show something really worry me – multi-node cluster performs worse than single node (due to inter-node communication I believe), and multi-cores deployment is slower than single-core deployment (this is something I don’t quite understand, may be because of L1/L2 cache?).

I need some hardware to test it as well, as VM is not that good for this kind of test. Then I suddenly recall I still have some 8-core/16G boxes in company sitting idle there, I can use one or two to do the test for sure ๐Ÿ˜‰

  3 Responses to “Cassandra and multi-core”

  1. Those powerful machines are running really old releases of all sorts of software, I tried some workaround to get things working, but the performance is not that exciting, will try some more tests.

  2. hello, how about the multicore experiment’s result?
    I have done similar experiments in single multicore node.
    I found the maximum concurrency is 3 for cassandra, and i don’t know why. If you are interested, can you contact me through email?

  3. Cannot recall all the details, but I’m pretty sure the multi-cores machines didn’t give me exciting results.

    Anyway, being away from Cassandra for quite something, it should have some improvement I believe, so you are on your own to do some more tests. ๐Ÿ˜‰

Sorry, the comment form is closed at this time.