On 5 Apr 2011, at 16:56, Steve Lloyd wrote:
> I believe so. (except I can't yet find the ATLAS numbers in the dashboard)
From http://dashb-atlas-job.cern.ch/dashboard/request.py/dailysummary make your usual selections of sites and activities. Then click the "CPU Efficiency" button.
The pie charts along the bottom have the CPU consumed (and the wallclock), with the usual links to other data formats, e.g.,
http://dashb-atlas-job.cern.ch/dashboard/request.py/consumptionscsv?sites=UK&activities=analysis&sitesSort=8&start=null&end=null&timeRange=lastWeek&granularity=Hourly&generic=1&sortBy=0&type=ewgb
gives the analysis raw CPU consumption per cluster in the last week.
Cheers
Graeme
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> + Steve Lloyd Queen Mary, University of London +
> + E-mail: [log in to unmask] School of Physics +
> + Phone: +44-(0)20-7882-5057 Mile End Road +
> + Fax: +44-(0)20-8981-9465 London E1 4NS, UK +
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>
>
>
>
> On 5 Apr 2011, at 16:46, Alessandra Forti wrote:
>
>> That's an unexpected twist quite elegant too. Would be it be:
>>
>> measured by ATLAS = RAW cpu
>> measured by APEL = HS06 hours
>>
>> ?
>>
>> cheers
>> alessandra
>>
>> On 05/04/2011 15:20, Steve Lloyd wrote:
>>> Hi,
>>> Can't we use something like:
>>>
>>> Analysis CPU (measured by ATLAS) * Total CPU (measured by APEL) / Total CPU (Measured by ATLAS)
>>> Production CPU (measured by ATLAS) * Total CPU (measured by APEL) / Total CPU (Measured by ATLAS)
>>>
>>> Cheers
>>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>> + Steve Lloyd Queen Mary, University of London +
>>> + E-mail: [log in to unmask] School of Physics +
>>> + Phone: +44-(0)20-7882-5057 Mile End Road +
>>> + Fax: +44-(0)20-8981-9465 London E1 4NS, UK +
>>> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>>
|