<div dir="ltr"><br><br><div class="gmail_quote">On Wed, Jun 16, 2010 at 4:29 PM, Oleg Goldshmidt <span dir="ltr"><<a href="mailto:pub@goldshmidt.org">pub@goldshmidt.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">
<div class="im">On Wed, Jun 16, 2010 at 1:23 PM, Shimon Panfil <<a href="mailto:info@industrialphys.com">info@industrialphys.com</a>> wrote:<br>
><br>
> Hi folks,<br>
> I'm looking for affordable workstation for heavy number crunching, not<br>
<br>
</div>What's "affordable" and what is "heavy number crunching"?<br>
<br>
For most large-scale scientific/engineering number-crunching physical<br>
parallelism (multiple CPUs/cores) is important for performance. Will<br>
you benefit from many more than 4 cores? Will anything more than<br>
commodity 4 core desktop be prohibitively expensive? Will you benefit<br>
/ can you afford, e.g., a CUDA-based number-cruncher under your desk?<br>
<div class="im"><br></div></blockquote><div><br></div><div>Note that while a CPU consumes around 100W at highest P-state (core voltage and frequency), based for example on opteron specifications, GPGPUs consume around 300W and even more when active. If the current box has a problem with dissipating heat, it will not take a GPGPU.</div>
<div><br></div><div>Furthermore, switching to GPGPU programming has a huge cost in labor. For this specific case, in which Shimon is a consultant, my assumption is that he works with a different piece of software every time, he will not have a chance to enjoy the fruit of his labor.</div>
</div><br clear="all"><br>-- <br>Orna Agmon Ben-Yehuda.<br><a href="http://ladypine.org">http://ladypine.org</a><br>
</div>