On Wed, Aug 06, 2014 at 01:33:36PM +0800, Ming Lei wrote: > With coroutine, the running time for same loading is increased > ~50%(1.325s vs. 0.903s), and dcache load events is increased
I agree with Paolo about microbenchmarks. We need to do I/O to get a realistic picture of performance, since there is little point is optimizing something that is not a significant factor in overall performance. But I also wanted to say that these benchmark durations are so short that they can be greatly affected by outliers (e.g. scheduler behavior, system background activity, etc). Run benchmarks for 2 minutes to reduce variance and give the system time to "warm up".
pgpdVv9EmCnOH.pgp
Description: PGP signature
