You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mahout.apache.org by Benson Margulies <bi...@gmail.com> on 2011/05/02 12:42:26 UTC

Microbenchmarks

Post 0.5, what do people think about populating some microbenchmarks?
If we're going to consider dumping collections for fastutil, it might
be good to have a measurement framework. I've had some success with
japex.

Re: Microbenchmarks

Posted by Dawid Weiss <da...@cs.put.poznan.pl>.
There is a new version of Google Caliper coming that will be very
interesting to watch. I temporarily froze JUnitBenchmarks waiting for the
new Caliper and I think I'll switch the internals once it is out (because it
makes little sense to duplicate measurement and reports infrastructure).

Dawid

On Mon, May 2, 2011 at 7:39 PM, Grant Ingersoll <gs...@apache.org> wrote:

>
> On May 2, 2011, at 12:53 PM, Benson Margulies wrote:
>
> > I'm aiming at microbenchmarks. No actually running on a cluster would
> > be involved. The idea is to measure the speeds of things like encoding
> > text vectors and matrix accesses and such.
> >
>
> Yes, we need both.
>
> > On Mon, May 2, 2011 at 12:14 PM, Oliver Fischer
> > <o....@swe-blog.net> wrote:
> >> Do you know about this issue. I am currently working on this, but I
> don't
> >> know when I will finish.
> >>
> >> Provide a performance measurement framework for Mahout
> >> => https://issues.apache.org/jira/browse/MAHOUT-670
> >>
> >> Oliver
> >>
> >>
> >>
> >> Am 02.05.2011 12:42, schrieb Benson Margulies:
> >>>
> >>> Post 0.5, what do people think about populating some microbenchmarks?
> >>> If we're going to consider dumping collections for fastutil, it might
> >>> be good to have a measurement framework. I've had some success with
> >>> japex.
> >>
> >> --
> >> Oliver B. Fischer, Schönhauser Allee 64, 10437 Berlin
> >> Certified ScrumMaster, OMG Certified Expert in BPM - Fundamental
> >> Tel. +49 30 44793251, Mobil: +49 178 7903538
> >> Mail: o.b.fischer@swe-blog.net
> >> Blog: http://logbuch.freiheitsgrade-se.de
> >>
>
>
>

Re: Microbenchmarks

Posted by Grant Ingersoll <gs...@apache.org>.
On May 2, 2011, at 12:53 PM, Benson Margulies wrote:

> I'm aiming at microbenchmarks. No actually running on a cluster would
> be involved. The idea is to measure the speeds of things like encoding
> text vectors and matrix accesses and such.
> 

Yes, we need both.  

> On Mon, May 2, 2011 at 12:14 PM, Oliver Fischer
> <o....@swe-blog.net> wrote:
>> Do you know about this issue. I am currently working on this, but I don't
>> know when I will finish.
>> 
>> Provide a performance measurement framework for Mahout
>> => https://issues.apache.org/jira/browse/MAHOUT-670
>> 
>> Oliver
>> 
>> 
>> 
>> Am 02.05.2011 12:42, schrieb Benson Margulies:
>>> 
>>> Post 0.5, what do people think about populating some microbenchmarks?
>>> If we're going to consider dumping collections for fastutil, it might
>>> be good to have a measurement framework. I've had some success with
>>> japex.
>> 
>> --
>> Oliver B. Fischer, Schönhauser Allee 64, 10437 Berlin
>> Certified ScrumMaster, OMG Certified Expert in BPM - Fundamental
>> Tel. +49 30 44793251, Mobil: +49 178 7903538
>> Mail: o.b.fischer@swe-blog.net
>> Blog: http://logbuch.freiheitsgrade-se.de
>> 



Re: Microbenchmarks

Posted by Benson Margulies <bi...@gmail.com>.
I'm aiming at microbenchmarks. No actually running on a cluster would
be involved. The idea is to measure the speeds of things like encoding
text vectors and matrix accesses and such.

On Mon, May 2, 2011 at 12:14 PM, Oliver Fischer
<o....@swe-blog.net> wrote:
> Do you know about this issue. I am currently working on this, but I don't
> know when I will finish.
>
> Provide a performance measurement framework for Mahout
> => https://issues.apache.org/jira/browse/MAHOUT-670
>
> Oliver
>
>
>
> Am 02.05.2011 12:42, schrieb Benson Margulies:
>>
>> Post 0.5, what do people think about populating some microbenchmarks?
>> If we're going to consider dumping collections for fastutil, it might
>> be good to have a measurement framework. I've had some success with
>> japex.
>
> --
> Oliver B. Fischer, Schönhauser Allee 64, 10437 Berlin
> Certified ScrumMaster, OMG Certified Expert in BPM - Fundamental
> Tel. +49 30 44793251, Mobil: +49 178 7903538
> Mail: o.b.fischer@swe-blog.net
> Blog: http://logbuch.freiheitsgrade-se.de
>

Re: Microbenchmarks

Posted by Oliver Fischer <o....@swe-blog.net>.
Do you know about this issue. I am currently working on this, but I 
don't know when I will finish.

Provide a performance measurement framework for Mahout
=> https://issues.apache.org/jira/browse/MAHOUT-670

Oliver



Am 02.05.2011 12:42, schrieb Benson Margulies:
> Post 0.5, what do people think about populating some microbenchmarks?
> If we're going to consider dumping collections for fastutil, it might
> be good to have a measurement framework. I've had some success with
> japex.

-- 
Oliver B. Fischer, Schönhauser Allee 64, 10437 Berlin
Certified ScrumMaster, OMG Certified Expert in BPM - Fundamental
Tel. +49 30 44793251, Mobil: +49 178 7903538
Mail: o.b.fischer@swe-blog.net
Blog: http://logbuch.freiheitsgrade-se.de

Re: Microbenchmarks

Posted by Grant Ingersoll <gs...@apache.org>.
On May 2, 2011, at 6:42 AM, Benson Margulies wrote:

> Post 0.5, what do people think about populating some microbenchmarks?
> If we're going to consider dumping collections for fastutil, it might
> be good to have a measurement framework. I've had some success with
> japex.


+1.  We definitely need to start benchmarking to get ready for 1.0.