<html><body><div style="color:#000; background-color:#fff; font-family:times new roman, new york, times, serif;font-size:16px"><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">Sorry about the false alarm re performance! After carefully looking at the numbers from the examples </div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">and versions etc, it turns out that the performance on Phi is lower than the host, which is understandable </div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">given the power and perf characteristics of single cores. As Wim had mentioned, the value of Phi will be </div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">visible if and when we can scale out. </div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr"><br></div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">Yury, </div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">I meant translated interpreter running on Phi v/s on the Xeon host, not translated v/s something else. </div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr"><br></div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">The device id is 0x225d so it looks like it's a Knights Corner 3120A. </div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr"><br></div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">Thanks, </div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">Ajit</div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr">==== </div><div id="yui_3_16_0_1_1429032627729_5076" dir="ltr"><br></div> <br><div class="qtdSeparateBR"><br><br></div><div class="yahoo_quoted" style="display: block;"> <div style="font-family: times new roman, new york, times, serif; font-size: 16px;"> <div style="font-family: HelveticaNeue, Helvetica Neue, Helvetica, Arial, Lucida Grande, sans-serif; font-size: 16px;"> <div dir="ltr"> <font size="2" face="Arial"> On Tuesday, April 14, 2015 9:24 AM, Yury V. Zaytsev <yury@shurup.com> wrote:<br> </font> </div> <blockquote style="border-left: 2px solid rgb(16, 16, 255); margin-left: 5px; margin-top: 5px; padding-left: 5px;"> <br><br> <div class="y_msg_container">On Mon, 2015-04-13 at 21:56 +0000, Ajit Dingankar wrote:<br clear="none">> Will post some results on performance; the translated interpreter<br clear="none">> seems much faster on Phi but I don't quite understand why! (Not that<br clear="none">> I'm complaining! ;-) <br clear="none"><br clear="none">Translated as compared to what ;-) ?<br clear="none"><br clear="none">If you compiled on the host with gcc (especially an older one) and now<br clear="none">comparing this to the cross-compiled version with icc, this will be very<br clear="none">unfair :-) Also, make sure you translate the same revision on both...<br clear="none"><br clear="none">I would be curious about recent results, and please be sure to mention<br clear="none">which Phi are you using for this test. I haven't had access to any<br clear="none">recent ones lately...<div class="yqt3913922052" id="yqtfd35924"><br clear="none"><br clear="none">Thanks,</div><br clear="none"><br clear="none">-- <br clear="none">Sincerely yours,<br clear="none">Yury V. Zaytsev<div class="yqt3913922052" id="yqtfd45692"><br clear="none"><br clear="none"><br clear="none"></div><br><br></div> </blockquote> </div> </div> </div></div></body></html>