<html><body><div style="color:#000; background-color:#fff; font-family:times new roman, new york, times, serif;font-size:12pt"><div><span>I run a few different distributed computing projects mostly 24/7 on a few quad core cpu's. While it's never happened to me I've heard plenty of stories about burning up cpu's but mostly in the old days where it took just seconds, especially if the heat sink was put on improperly or forgotten altogether. It seems to be somewhat less of an issue now with lower voltage cpu's, improved heat sinks, better ventilated cases, and thermal monitoring in bios where yes, it many circumstances the default setting was no monitoring.</span></div><div style="color: rgb(0, 0, 0); font-size: 16px; font-family: 'times new roman', 'new york', times, serif; background-color: transparent; font-style: normal;"><span><br></span></div><div style="color: rgb(0, 0, 0); font-size: 16px; font-family: 'times new roman', 'new york',
times, serif; background-color: transparent; font-style: normal;"><span>I monitor temps with the command 'sensors' as well as running the gui equivalent xsensors. When the temps go above 58C in the winter it's usually a sign of dust build up so out comes the air compressor for a good blasting. Ambients are just too warm during the summer to run fun tilt unless the cpu's are clocked way back.</span></div><div style="color: rgb(0, 0, 0); font-size: 16px; font-family: 'times new roman', 'new york', times, serif; background-color: transparent; font-style: normal;"><span><br></span></div><div style="color: rgb(0, 0, 0); font-size: 16px; font-family: 'times new roman', 'new york', times, serif; background-color: transparent; font-style: normal;"><span>A lot of people are using gpu's these days to run projects. I've heard the failure rate is pretty high as vid cards simply aren't designed to run high loads for such long durations.
</span></div><div><br></div> <div style="font-family: 'times new roman', 'new york', times, serif; font-size: 12pt;"> <div style="font-family: 'times new roman', 'new york', times, serif; font-size: 12pt;"> <div dir="ltr"> <font size="2" face="Arial"> <hr size="1"> <b><span style="font-weight:bold;">From:</span></b> Bob Kline <bob.kline@gmail.com><br> <b><span style="font-weight: bold;">To:</span></b> "Mailing List for LUG in greater Grand Rapids, MI area." <grlug@grlug.org> <br> <b><span style="font-weight: bold;">Sent:</span></b> Friday, December 7, 2012 12:18 PM<br> <b><span style="font-weight: bold;">Subject:</span></b> [GRLUG] Burn up the CPU - NOT LINUX<br> </font> </div> <br>
<div id="yiv841512478">On a scale of folklore to fact, is <div>there anything to the notion that </div><div>one can burn up a CPU by keeping</div><div>it at full throttle for an extended </div><div>period?</div><div><br></div><div>Consider just nominal clock rates,</div>
<div>not over clocking.</div><div><br></div><div>e.g., if someone, me for example,</div><div>was to process SETI data 24/7, is</div><div>the CPU likely to smoke?</div><div><br></div><div>I can imagine that a particularly </div>
<div>dusty CPU and cooling fin might</div><div>pose issues. Anything else? Even</div><div>that?</div><div><br></div><div>Some motherboards have thermal</div><div>protection of various kinds. Is there</div><div>an open software package that can</div>
<div>display where things are at? In the</div><div>olden days I think a motherboard </div><div>would sometimes come with a DOS</div><div>or windoz package that would do that.</div><div><br></div><div> -- Bob</div><div>
<br></div>
</div><br>_______________________________________________<br>grlug mailing list<br><a ymailto="mailto:grlug@grlug.org" href="mailto:grlug@grlug.org">grlug@grlug.org</a><br><a href="http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug" target="_blank">http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug</a><br><br> </div> </div> </div></body></html>