performance - How is 6M TPS counted/measured in LMAX's disruptor pattern? -


What is the number of factors in the IO operation? Or is it just the number of transactions, can the architecture read and match with memory / writing?

What hardware do they use during testing? It probably does not matter, but what OS used it?

With LMAX, IO usually has transaction loop and elsewhere / P>

As long as no one can answer that in order to actually run that benchmark, I can only assume that they have some kind of counter and any type of time measurement. But I do not think it was on a multi-socket configuration.

In addition to this 6 M TPS hearing from 2011 or 2010, you should run your test with a dispenser. The benchmarks they wanted to run could not reflect either that you need to complete, and, unfortunately, this latency / throughput number will vary greatly for slight changes in the circumstances. Knowing the OS or hardware can cure your curiosity, but this number will not make more meaningful than your own evaluation.

Comments

Popular posts from this blog

Pass DB Connection parameters to a Kettle a.k.a PDI table Input step dynamically from Excel -

multithreading - PhantomJS-Node in a for Loop -

c++ - MATLAB .m file to .mex file using Matlab Compiler -