Lux, Jim (337C)
james.p.lux at jpl.nasa.gov
Mon Nov 5 09:12:48 PST 2012
On 11/5/12 8:50 AM, "Douglas Eadline" <deadline at eadline.org> wrote:
>>>More interesting is the ECC discussion.
>>>ECC is simply a requirement IMHO, not a 'luxury thing' as some
>>>hardware engineers see it.
>> Depends on your computational model. Would you rather spend money on
>> or on more processors?
>> ECC comes at a cost in speed as well. There is some non-zero time
>> required to compute the syndrome bits and do the correction on the read.
>> Sure, you can pipeline it, but there's some extra latency inevitably
>I find it interesting that many users thought GPU's could not be
>a research tool unless they had ECC memory. I have one associate who
>turns it off because they get 10% better performance on their
Assuming one can tell if one's simulation results are "good", this is a
More information about the Beowulf