.... and also how many executing jobs/core.
> -----Original Message-----
> From: Testbed Support for GridPP member institutes [mailto:TB-
> [log in to unmask]] On Behalf Of Greig A. Cowan
> Sent: 25 November 2008 15:17
> To: [log in to unmask]
> Subject: Re: Analysis challenge in the UK tomorrow
>
> Ahh, you are right. I hadn't realised the monitoring was that
detailed.
> If you look at the plots for the UK sites with 20 jobs/site you will
> see
> that on average, the jobs were 61.5% efficient. There is quite a
spread
> of efficiency depending on the site. It would be good to compare
> network
> topologies and storage hardware for each of these sites.
>
> Greig
>
>
>
> Gordon, JC (John) wrote, On 25/11/08 14:41:
> > ..... or rather, it would if there were any data there:-(
> >
> >
> >> -----Original Message-----
> >> From: Gordon, JC (John)
> >> Sent: 25 November 2008 14:40
> >> To: [log in to unmask]
> >> Subject: RE: Analysis challenge in the UK tomorrow
> >>
> >> Greig, the gangarobot link Graeme sent shows job efficiency plots
> per
> >> site.
> >>
n number SC005336.
--
Scanned by iCritical.
|