On Dec 17, 2012, at 11:23 PM, Mark Hahn wrote:
>> "todays 1 gflop/watt" ?
>
> press releases always put the new shiny thing in the best light.
> they're probably thinking of a conventional compute node,
> (say, 32 cores, 2.3 GHz, 4 flops/cycle, or 16c and 8 f/c -
> either way totalling 294 Gflops
It could be referred to DP (Dual Precision) GFlops/W, not single.
On 12/17/2012 02:50 PM, Vincent Diepeveen wrote:
> "todays 1 gflop/watt" ?
>
> The K20X delivers 1.4 Tflop nearly.
> If i google it's 235 watt TDP.
>
> 1.4 Tflop / 235 = 6 gflops/watt
>
> On Dec 17, 2012, at 2:21 PM, Eugen Leitl wr
On Thu, Dec 6, 2012 at 12:53 PM, Chi Chan wrote:
> On Wed, Nov 28, 2012 at 12:10 PM, Rayson Ho wrote:
>> 1) We ran a 10,000-node cluster on Amazon EC2 for Grid Engine
>> scalability testing a few weeks ago:
>>
>> http://blogs.scalablelogic.com/2012/11/running-1-node-grid-engine-cluster.html
>
> "todays 1 gflop/watt" ?
press releases always put the new shiny thing in the best light.
they're probably thinking of a conventional compute node,
(say, 32 cores, 2.3 GHz, 4 flops/cycle, or 16c and 8 f/c -
either way totalling 294 Gflops for 300W or less.)
> The K20X delivers 1.4 Tflop nearly.
I wasn't thinking so much about code efficiency, more "wall plug power"
efficiency. The board may consume 250W, but it will take non-zero power to
support that board, and then the power supply efficiency needs to be taken into
account. But I suspect the 1 GFLOP/W was more just an "old" "rounde
On Dec 17, 2012, at 8:15 PM, Lux, Jim (337C) wrote:
> I wasn't thinking so much about code efficiency, more "wall plug
> power" efficiency. The board may consume 250W, but it will take
> non-zero power to support that board, and then the power supply
> efficiency needs to be taken into acc
That could be a notional 1 GFLOP/Watt in a fielded system. The original
documents for PERFECT are probably a year or two old by now.. but what DARPA is
looking for is a nearly 2 order of magnitude improvement... Whether they
started at 1 or 1.4 or 6 really doesn't make much difference to what
On Dec 17, 2012, at 6:27 PM, Lux, Jim (337C) wrote:
> That could be a notional 1 GFLOP/Watt in a fielded system.
Even linpack is 70% - 80% efficient on this so should get out oh
let's use a conservative 4.5 flops/watt effectively at codes.
Note that (to my big surprise) it seems to be the case
On Sat, Dec 15, 2012 at 12:31 PM, Eric Lee wrote:
> As I mentioned in the thread I would like to run some bioinformatics
> applications on it. And using OGS as DRM.
We ported Grid Engine (ie. OGS/GE) to ARM Linux in the GE 2011.11
release, but we haven't encountered huge demands yet. May be when
"todays 1 gflop/watt" ?
The K20X delivers 1.4 Tflop nearly.
If i google it's 235 watt TDP.
1.4 Tflop / 235 = 6 gflops/watt
On Dec 17, 2012, at 2:21 PM, Eugen Leitl wrote:
>
> http://www.networkworld.com/community/blog/darpa-awards-20m-nvidia-
> stretch-achilles-heel-advanced-computing-power
>
http://www.networkworld.com/community/blog/darpa-awards-20m-nvidia-stretch-achilles-heel-advanced-computing-power
DARPA awards $20M to Nvidia to stretch "Achilles Heel" of advanced computing:
Power
Processor-maker Nvidia will get chance to boost power chip output from 1
GFLOPS/watt to 75 GFLOPS
>> the seq stuff I see is quite IO-intensive, sometimes memory-intensive.
> How do I know if the application is IO-intensive or memory-intensive? Any
> tools to measure that?
code inspection is the best tool. you should know the number and size of
files your workflow touches, and how it accesse
-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
On 14/12/12 11:10, Lux, Jim (337C) wrote:
> While the RPI might only be $35, you also have to buy a SD card
> ($10-12) power supplies, cables, boxes or mounting hardware, not to
> mention a network switch, etc.
Or a group of people could each send a
On 2012/12/17 12:30, Mark Hahn wrote:
>> Thanks for the information. I would like to set up RPi cluster for some
>> bioinformatics jobs like sequences alignment and sequences operations.
>
> you really need to look at where your code is spending its time.
> the seq stuff I see is quite IO-intensiv
On 2012/12/17 12:30, Mark Hahn wrote:
>> Thanks for the information. I would like to set up RPi cluster for some
>> bioinformatics jobs like sequences alignment and sequences operations.
>
> you really need to look at where your code is spending its time.
> the seq stuff I see is quite IO-intensive
15 matches
Mail list logo
