Skip to content

Instantly share code, notes, and snippets.

@PharkMillups
Created August 2, 2010 17:43
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save PharkMillups/505008 to your computer and use it in GitHub Desktop.
Save PharkMillups/505008 to your computer and use it in GitHub Desktop.
0:55 <vicmargar> Hello, what is the best way to use the protocol buffers client?
10:55 <vicmargar> If I have thousands of processes that want to access riak,
should I have a pool of connections?
12:06 <rtilder> vicmargar: Generally speaking, yes.
12:07 <vicmargar> what do you mean?
12:15 <vicmargar> I've been running some benchmarks with basho_bench and I see that
when I increase the number of concurrent workers the CPU usage of the machines increseases.
12:16 <vicmargar> So I'm assuming that I'll have to make the same number of concurrent
requests from my code to get similar performance to what I got in the test.
12:20 <benblack> yes.
12:21 <benblack> you will not see maximum performance on a single connection.
12:29 <vicmargar> benblack: yes but I guess also I have to make sure that the number
of connections distributed across all my nodes to riak isn't higher than the optimal maximum
12:30 <vicmargar> I guess it's just a matter of testing different combinations
and seeing what works better.
12:30 <benblack> correct
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment