Parallella Community Supercomputing for Everyone 2015-10-01T20:33:19+00:00 https://parallella.org/forums/feed.php?f=23&t=3222 2015-10-01T20:33:19+00:00 2015-10-01T20:33:19+00:00 https://parallella.org/forums/viewtopic.php?t=3222&p=15916#p15916 <![CDATA[Re: Why is one eCore faster then the Arm?]]>
etim wrote:
I did a little benchmark today running this code on a single eCore:
...
Can anyone explain why the eCore is faster at this benchmark?

I suggest you take a look at the generated assembly code, the Epiphany can do a multiply & add in one cycle, I don't think the basic ARM core can do that. Maybe the NEON cores can but I expect the code you wrote won't be executed on the NEON.

-jsg

Statistics: Posted by justsomeguy — Thu Oct 01, 2015 8:33 pm

]]>
2015-10-01T19:59:30+00:00 2015-10-01T19:59:30+00:00 https://parallella.org/forums/viewtopic.php?t=3222&p=15915#p15915 <![CDATA[Why is one eCore faster then the Arm?]]>
Code:
`for (n = 0; n < TIMES; n++){    //Clear Sum  (*(c))=0x0;  //Sum of product calculation  for (i = 0; i < N/CORES; i++){    (*(c)) += a[i] * b[i];  }}`

Then I did the same on the ARM:

Code:
`for (n = 0; n < TIMES; n++){    // printf("j= %d\n", j);    //Clear Sum    sop = 0;    //Sum of product calculation    for (i = 0; i < N; i++){      sop += a[i] * b[i];    }  }`

For TIMES=100,000 and N=4096, the eCore takes 11 seconds and the ARM takes 19 seconds.

Can anyone explain why the eCore is faster at this benchmark?

Statistics: Posted by etim — Thu Oct 01, 2015 7:59 pm

]]>