[Beowulf] New HPCC results and the Myri viewpoint
diep at xs4all.nl
Wed Jul 20 20:07:08 PDT 2005
>> Tests at all processors at the same time make major sense.
>Yes and no. Most networking people believe the job of a node is to send
>messages. Actually, it's mainly to compute, and sometimes sends
>messages. So, would running a pingpong test on multiple processors at
>the same time sharing a NIC an interesting benchmark ? Not really, it
>won't happen much on real codes that compute most of the time. I prefer
>to optimize other things that help the host compute faster.
If most of the time they are 'just computing', then it just doesn't make
sense to have a highend network. A $10 gigabit network will do in that case.
Reality is however different. Reality is that you simply stress the network
until it wastes say 10-20% of your system time until a maximum of 50%.
50% scaling of applications is acceptable, if it exponential speeds up
100% embarrassingly parallel calculations you can so to speak make a
distributed project for with 1 internet server and distribute software.
In short, if you deliver highend nic's, ASSUME they get used.
At least *i* understand that principle.
Weirdly enough the manufacturer of a product assumes his stuff isn't going
to get used.
Why make it then for your users?
You try to sell a product without your users using it?
>> Any denial in advance that it will be the same speed is just ballony.
>And I thought I was the bulliest on this list...
>I just give my opinion and at least my opinion is backed up by
>first-hand experience. I don't know how to play chess, but I know my stuff.
More information about the Beowulf