hi Amjad, I would like to point to 1 really important practical difference.
co-processors can have magnificent latency from the co-processor to the cpu. In the nano-seconds. Accelerators i'd argue typically go via pci-e or similar, so have horrible latency, in the microseconds or slower. Accelerators are just there for massive bandwidth based number crunching in general - take Xeon Phi or K20 from Nvidia or AMD's attempt there. On Mar 10, 2013, at 5:36 PM, amjad ali wrote: > Hello All, > > Is there any line/point to make distinction between accelerators > and co-processors (that are used in conjunction with the primary > CPU to boost up the performance)? or these terms can be used > interchangeably? > > Specifically, the word "accelerator" is used commonly with GPU. On > the other hand the word "co-processors" is used commonly with Xeon > Phi. > > Thank you. > > Regards, > Amjad > _______________________________________________ > Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin > Computing > To change your subscription (digest mode or unsubscribe) visit > http://www.beowulf.org/mailman/listinfo/beowulf _______________________________________________ Beowulf mailing list, Beowulf@beowulf.org sponsored by Penguin Computing To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
