Geek Answers Handbook

Cassandra proprietary binary protocol performance

I am currently planning to move a fairly large MongoDB deployment (~ 25 large servers located across multiple data centers) to Kassandra. I saw that Cassandra 1.2 has a relatively new proprietary binary protocol, and it seems that this is the future of the project. It was good when I read about it, but I recently tested it against a simple Trrift-based client, and the Thrift client exceeded it significantly in every test. My question is, in which scenario does the native binary protocol / client outperform Thrift and does anyone use it in production?

Here is an example:

  Local 3 node cluster (created using ccm), with 10 client threads spamming it
 Total throughput (write, read) [total ops in 5 minutes]:
 Binary: 3,316,725.0 2,179,193.0 
 Thrift: 7,310,482.0 2,872,528.0
+6
source share
1 answer

The binary protocol is not used when you are using an existing client like Astyanax or Hector. You can take a look at java-driver . It has not yet been released, but it works correctly as far as I tested.

+1
source

More articles:


All Articles