 |
 |
Hi, Was also briefly covered in the HS06 talk at HEPiX. http://indico.cern.ch/getFile.py/access?contribId=39&sessionId=5&resId=1&materialId=slides&confId=247864 Chris. > -----Original Message----- > From: Testbed Support for GridPP member institutes [mailto:TB- > [log in to unmask]] On Behalf Of Sam Skipsey > Sent: 07 November 2013 10:19 > To: [log in to unmask] > Subject: Re: CMS ARM CPU > > I think I saw a thing about this at CHEP, too (although all I can seem to find > post-hoc are Peter Elmer's rather preliminary investigations for CMS, which > have nice data but not enough experience to make a firm recommendation - > http://indico.cern.ch/getFile.py/access?contribId=303&sessionId=7&resId=0 > &materialId=slides&confId=214784 > ) > > Of course, the issue is always code recompilation... > Sam > > On 7 November 2013 08:45, Jensen, Jens (STFC,RAL,SC) > <[log in to unmask]> wrote: > > This may be of interest: http://xxx.soton.ac.uk/abs/1311.0269 > > > > The premise is to go "greener" by running stuff on ARM CPUs. They get > > fewer events per minute per core than on Xeons, but more events per > > minute per Watt. > > > > -j > > > > -- > > Scanned by iCritical. -- Scanned by iCritical.