Re: [eigen] Help needed to run a benchmark on many machines

[Benoit Jacob <jacob.benoit.1@xxxxxxxxx>]

2015-02-18 15:47 GMT-05:00 Hauke Heibel <hauke.heibel@xxxxxxxxx>:

On Wed, Feb 18, 2015 at 9:37 PM, Benoit Jacob <jacob.benoit.1@xxxxxxxxx> wrote:
>> c) Do we actually need to test user CPU times as opposed to wall times?
>
>
> Probably not - do you have an opinion on this?

Not really, my (most likely) naive assumption was that user cpu times
might reduce noise (caused e.g. by OS thread scheduling on system with
multiple cores and hyperthreading).

Same here. So when I can easily get cpu time, I use it, and otherwise (on Mac) i just use gettimeofday and (given enough repetitions of measurements) this works well enough.

Benoit