<div dir="ltr">Thank you Nikhil, the nkp option works pretty well!<br></div><div class="gmail_extra"><br><div class="gmail_quote">On Mon, Jun 5, 2017 at 10:13 AM, Nikhil Jain <span dir="ltr"><<a href="mailto:nikhil.jain@acm.org" target="_blank">nikhil.jain@acm.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div style="word-wrap:break-word">I don’t think 10 microsecond is a large lookahead; How big a system are you trying to simulate (as in the network size)? If the system is small, it may be hard to reduce the rollbacks as the #core increase.<div><br></div><div>Other things that can be tried:</div><div>1. nkp option = number of KPs; but default ROSS uses 16, but you can set it to (#total LPs)/(#cores) - i.e. make each LP has its own KP, which will prevent false rollbacks</div><div>2. gvt/batch size = you can try a small gvt frequency (like 4/8) and batch size (4/8).</div><div><br></div><div><div>
<div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;word-wrap:break-word"><div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;word-wrap:break-word"><div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;word-wrap:break-word"><div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;word-wrap:break-word">---</div><div style="color:rgb(0,0,0);letter-spacing:normal;text-align:start;text-indent:0px;text-transform:none;white-space:normal;word-spacing:0px;word-wrap:break-word">Nikhil Jain<br>Postdoctoral Fellow, Lawrence Livermore National Laboratory<br><a href="mailto:nikhil.jain@acm.org" target="_blank">nikhil.jain@acm.org</a>, <a href="http://nikhil-jain.github.io/" target="_blank">http://nikhil-jain.github.io/</a></div></div></div></div>
</div>
<br><div><blockquote type="cite"><div><div class="h5"><div>On Jun 5, 2017, at 11:57, Jian Peng <<a href="mailto:jpeng10@hawk.iit.edu" target="_blank">jpeng10@hawk.iit.edu</a>> wrote:</div><br class="m_-4524295584132203420Apple-interchange-newline"></div></div><div><div><div class="h5"><div dir="ltr"><div><div>Hi,<br><br></div> My simulation is having too many events rolled back, like 70%. It's still relatively small scale, 16 cores on 4 nodes, using optimistic scheduler. The result is attached.<br><br> Any general idea of reducing number of rollback events? I used the max-opt-lookahead option and set it to 10000. Is it still a large number?<br><br></div> Thanks!<br></div>
</div></div><span id="m_-4524295584132203420cid:2F4A1176-21FB-4053-9408-F4DC07F681A2@ka.ltv"><rollback_result_1.png></span><span id="m_-4524295584132203420cid:0E0E128C-4B6F-40AD-9310-638784EFFE96@ka.ltv"><<wbr>rollback_result_2.png></span>________<wbr>______________________________<wbr>_________<br>codes-ross-users mailing list<br><a href="mailto:codes-ross-users@lists.mcs.anl.gov" target="_blank">codes-ross-users@lists.mcs.<wbr>anl.gov</a><br><a href="https://lists.mcs.anl.gov/mailman/listinfo/codes-ross-users" target="_blank">https://lists.mcs.anl.gov/<wbr>mailman/listinfo/codes-ross-<wbr>users</a><br></div></blockquote></div><br></div></div></blockquote></div><br></div>