<div dir="ltr">Perhaps you can try to set the IPOPT option 'print_timing_statistics' = 'yes': <a href="https://www.coin-or.org/Ipopt/documentation/node41.html">https://www.coin-or.org/Ipopt/documentation/node41.html</a></div><div class="gmail_extra"><br><div class="gmail_quote">On Wed, Jan 18, 2017 at 10:00 AM, <span dir="ltr"><<a href="mailto:ipopt-request@coin-or.org" target="_blank">ipopt-request@coin-or.org</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Send Ipopt mailing list submissions to<br>
<a href="mailto:ipopt@list.coin-or.org">ipopt@list.coin-or.org</a><br>
<br>
To subscribe or unsubscribe via the World Wide Web, visit<br>
<a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__list.coin-2Dor.org_mailman_listinfo_ipopt&d=CwICAg&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=BAP9D2FFNY-E7nYzFI5fHw&m=sdB5SWDdjesA3HFY6oMVN57SUQ3e1bD-NFPtNrbSvys&s=y0CInmYX6aVEEeJB8dgOXO9lBv0UNk6v33OORxrQTLo&e=" rel="noreferrer" target="_blank">https://urldefense.proofpoint.<wbr>com/v2/url?u=http-3A__list.<wbr>coin-2Dor.org_mailman_<wbr>listinfo_ipopt&d=CwICAg&c=Ngd-<wbr>ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLx<wbr>WPA_2Wlc4&r=BAP9D2FFNY-<wbr>E7nYzFI5fHw&m=<wbr>sdB5SWDdjesA3HFY6oMVN57SUQ3e1b<wbr>D-NFPtNrbSvys&s=<wbr>y0CInmYX6aVEEeJB8dgOXO9lBv0UNk<wbr>6v33OORxrQTLo&e=</a><br>
or, via email, send a message with subject or body 'help' to<br>
<a href="mailto:ipopt-request@list.coin-or.org">ipopt-request@list.coin-or.org</a><br>
<br>
You can reach the person managing the list at<br>
<a href="mailto:ipopt-owner@list.coin-or.org">ipopt-owner@list.coin-or.org</a><br>
<br>
When replying, please edit your Subject line so it is more specific<br>
than "Re: Contents of Ipopt digest..."<br>
<br>
<br>
Today's Topics:<br>
<br>
1. WSMP for IPOPT (Stuart Rogers)<br>
2. Getting Accurate CPU Times from Ipopt (Anil Rao)<br>
<br>
<br>
------------------------------<wbr>------------------------------<wbr>----------<br>
<br>
Message: 1<br>
Date: Tue, 17 Jan 2017 13:37:39 -0700<br>
From: Stuart Rogers <<a href="mailto:smr1@ualberta.ca">smr1@ualberta.ca</a>><br>
To: <a href="mailto:ipopt@list.coin-or.org">ipopt@list.coin-or.org</a><br>
Subject: [Ipopt] WSMP for IPOPT<br>
Message-ID:<br>
<CAHBg-_oNP4X5vCtig7RcrZuys=<wbr>NG2cXg6vZK2Pkeyi4q=<a href="mailto:uEm0w@mail.gmail.com">uEm0w@mail.<wbr>gmail.com</a>><br>
Content-Type: text/plain; charset="utf-8"<br>
<br>
I have a few questions regarding WSMP and how it is used by IPOPT.<br>
<br>
* The WSMP webpage (<a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__researcher.watson.ibm&d=CwICAg&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=BAP9D2FFNY-E7nYzFI5fHw&m=sdB5SWDdjesA3HFY6oMVN57SUQ3e1bD-NFPtNrbSvys&s=Iz9GXtbtQBWDczHVOY8Aabhixqy90l_2D6hyVtuUesQ&e=" rel="noreferrer" target="_blank">https://urldefense.<wbr>proofpoint.com/v2/url?u=http-<wbr>3A__researcher.watson.ibm&d=<wbr>CwICAg&c=Ngd-<wbr>ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLx<wbr>WPA_2Wlc4&r=BAP9D2FFNY-<wbr>E7nYzFI5fHw&m=<wbr>sdB5SWDdjesA3HFY6oMVN57SUQ3e1b<wbr>D-NFPtNrbSvys&s=<wbr>Iz9GXtbtQBWDczHVOY8Aabhixqy90l<wbr>_2D6hyVtuUesQ&e=</a> .<br>
com/researcher/view_group.php?<wbr>id=1426) says that version 15 includes<br>
GPU-accelerated libraries. Which WSMP solvers (i.e. symmetric direct,<br>
general direct, or iterative solvers) have been implemented to run on the<br>
GPUs? Was the GPU acceleration of WSMP implemented in CUDA or OpenCL? Are<br>
there any benchmark results showing a marked speed-up of IPOPT when using<br>
the GPU-accelerated version of WSMP?<br>
<br>
* The WSMP webpage seems to only offer downloads of WSMP. Is PWSMP included<br>
in these downloads? Is IPOPT able to utilize the PWSMP linear solver? If<br>
so, are there any benchmark results showing a marked speed-up of IPOPT when<br>
using PWSMP rather than WSMP?<br>
<br>
* The default value in IPOPT for the wsmp_num_threads option is 1:<br>
<a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__www.coin-2Dor.org_Ipopt_documentation_node62.html&d=CwICAg&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=BAP9D2FFNY-E7nYzFI5fHw&m=sdB5SWDdjesA3HFY6oMVN57SUQ3e1bD-NFPtNrbSvys&s=OzkYEyXCFXY-2nquX-ehE6Qitij6QQqxSWfwhmuJGig&e=" rel="noreferrer" target="_blank">https://urldefense.proofpoint.<wbr>com/v2/url?u=https-3A__www.<wbr>coin-2Dor.org_Ipopt_<wbr>documentation_node62.html&d=<wbr>CwICAg&c=Ngd-<wbr>ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLx<wbr>WPA_2Wlc4&r=BAP9D2FFNY-<wbr>E7nYzFI5fHw&m=<wbr>sdB5SWDdjesA3HFY6oMVN57SUQ3e1b<wbr>D-NFPtNrbSvys&s=OzkYEyXCFXY-<wbr>2nquX-<wbr>ehE6Qitij6QQqxSWfwhmuJGig&e=</a> . This option<br>
determines the number of threads to be used by WSMP. In the case when the<br>
user doesn't set this option, why doesn't IPOPT or WSMP internally set this<br>
option to the number of physical cores detected rather than defaulting to 1?<br>
<br>
* wsmp_iterative is an IPOPT option that uses the iterative rather than<br>
symmetric direct WSMP solver:<br>
<a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__www.coin-2Dor.org_Bonmin_option-5Fpages_options-5Flist-5Fipopt.html&d=CwICAg&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=BAP9D2FFNY-E7nYzFI5fHw&m=sdB5SWDdjesA3HFY6oMVN57SUQ3e1bD-NFPtNrbSvys&s=GlF-iE9dbvi-t1mjuWs24cuv0aFUe_AXDb2PVDJmoeA&e=" rel="noreferrer" target="_blank">https://urldefense.proofpoint.<wbr>com/v2/url?u=https-3A__www.<wbr>coin-2Dor.org_Bonmin_option-<wbr>5Fpages_options-5Flist-<wbr>5Fipopt.html&d=CwICAg&c=Ngd-<wbr>ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLx<wbr>WPA_2Wlc4&r=BAP9D2FFNY-<wbr>E7nYzFI5fHw&m=<wbr>sdB5SWDdjesA3HFY6oMVN57SUQ3e1b<wbr>D-NFPtNrbSvys&s=GlF-iE9dbvi-<wbr>t1mjuWs24cuv0aFUe_<wbr>AXDb2PVDJmoeA&e=</a> . Are<br>
there any benchmark results showing a marked speed-up of IPOPT when using<br>
this iterative WSMP option?<br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <<a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__list.coin-2Dor.org_pipermail_ipopt_attachments_20170117_512fbcec_attachment-2D0001.html&d=CwICAg&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=BAP9D2FFNY-E7nYzFI5fHw&m=sdB5SWDdjesA3HFY6oMVN57SUQ3e1bD-NFPtNrbSvys&s=IA2HD8kHt605poyEJzfgla_HVFCofAkuQL-2CVgHX3o&e=" rel="noreferrer" target="_blank">https://urldefense.<wbr>proofpoint.com/v2/url?u=http-<wbr>3A__list.coin-2Dor.org_<wbr>pipermail_ipopt_attachments_<wbr>20170117_512fbcec_attachment-<wbr>2D0001.html&d=CwICAg&c=Ngd-<wbr>ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLx<wbr>WPA_2Wlc4&r=BAP9D2FFNY-<wbr>E7nYzFI5fHw&m=<wbr>sdB5SWDdjesA3HFY6oMVN57SUQ3e1b<wbr>D-NFPtNrbSvys&s=<wbr>IA2HD8kHt605poyEJzfgla_<wbr>HVFCofAkuQL-2CVgHX3o&e=</a> ><br>
<br>
------------------------------<br>
<br>
Message: 2<br>
Date: Tue, 17 Jan 2017 15:41:19 -0500<br>
From: Anil Rao <<a href="mailto:anilvrao@gmail.com">anilvrao@gmail.com</a>><br>
To: <a href="mailto:ipopt@list.coin-or.org">ipopt@list.coin-or.org</a><br>
Subject: [Ipopt] Getting Accurate CPU Times from Ipopt<br>
Message-ID: <<a href="mailto:8284AB55-1A48-42F5-B636-B2D63F3C1D1C@gmail.com">8284AB55-1A48-42F5-B636-<wbr>B2D63F3C1D1C@gmail.com</a>><br>
Content-Type: text/plain; charset="us-ascii"<br>
<br>
I would like to know how to get the most accurate CPU time from Ipopt. I am running a large set of jobs and I need the CPU time from each job. I am using the MATLAB mex files compiled by Enrico Bertolazzi for my runs. I am finding that Ipopt does not seem to provide accurate CPU times using the CPU time output from the mex interface. I am grateful if somebody might be able to help me get an accurate execution time. Thanks for your help.<br>
<br>
Regards,<br>
<br>
Anil<br>
<br>
Sent from my iPhone. Please excuse any typos.<br>
<br>
Anil V. Rao, PhD<br>
Gainesville, FL 32607<br>
Tel: <a href="tel:352-672-1529" value="+13526721529">352-672-1529</a> (cell)<br>
Email: <a href="mailto:anilvrao@gmail.com">anilvrao@gmail.com</a><br>
Website: <a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__www.anilvrao.com&d=CwICAg&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=BAP9D2FFNY-E7nYzFI5fHw&m=sdB5SWDdjesA3HFY6oMVN57SUQ3e1bD-NFPtNrbSvys&s=mC1QomAhf4b9A5pWtDOkca9s3pFmzVMjG4NPN_Krizg&e=" rel="noreferrer" target="_blank">https://urldefense.proofpoint.<wbr>com/v2/url?u=http-3A__www.<wbr>anilvrao.com&d=CwICAg&c=Ngd-<wbr>ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLx<wbr>WPA_2Wlc4&r=BAP9D2FFNY-<wbr>E7nYzFI5fHw&m=<wbr>sdB5SWDdjesA3HFY6oMVN57SUQ3e1b<wbr>D-NFPtNrbSvys&s=<wbr>mC1QomAhf4b9A5pWtDOkca9s3pFmzV<wbr>MjG4NPN_Krizg&e=</a><br>
-------------- next part --------------<br>
An HTML attachment was scrubbed...<br>
URL: <<a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__list.coin-2Dor.org_pipermail_ipopt_attachments_20170117_6c682c6f_attachment-2D0001.html&d=CwICAg&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=BAP9D2FFNY-E7nYzFI5fHw&m=sdB5SWDdjesA3HFY6oMVN57SUQ3e1bD-NFPtNrbSvys&s=N4m7utDvSU8ZmHmXkeTkVLXjr1Z3AuE7sGDUqLVqCII&e=" rel="noreferrer" target="_blank">https://urldefense.<wbr>proofpoint.com/v2/url?u=http-<wbr>3A__list.coin-2Dor.org_<wbr>pipermail_ipopt_attachments_<wbr>20170117_6c682c6f_attachment-<wbr>2D0001.html&d=CwICAg&c=Ngd-<wbr>ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLx<wbr>WPA_2Wlc4&r=BAP9D2FFNY-<wbr>E7nYzFI5fHw&m=<wbr>sdB5SWDdjesA3HFY6oMVN57SUQ3e1b<wbr>D-NFPtNrbSvys&s=<wbr>N4m7utDvSU8ZmHmXkeTkVLXjr1Z3Au<wbr>E7sGDUqLVqCII&e=</a> ><br>
<br>
------------------------------<br>
<br>
Subject: Digest Footer<br>
<br>
______________________________<wbr>_________________<br>
Ipopt mailing list<br>
<a href="mailto:Ipopt@list.coin-or.org">Ipopt@list.coin-or.org</a><br>
<a href="https://urldefense.proofpoint.com/v2/url?u=http-3A__list.coin-2Dor.org_mailman_listinfo_ipopt&d=CwICAg&c=Ngd-ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLxWPA_2Wlc4&r=BAP9D2FFNY-E7nYzFI5fHw&m=sdB5SWDdjesA3HFY6oMVN57SUQ3e1bD-NFPtNrbSvys&s=y0CInmYX6aVEEeJB8dgOXO9lBv0UNk6v33OORxrQTLo&e=" rel="noreferrer" target="_blank">https://urldefense.proofpoint.<wbr>com/v2/url?u=http-3A__list.<wbr>coin-2Dor.org_mailman_<wbr>listinfo_ipopt&d=CwICAg&c=Ngd-<wbr>ta5yRYsqeUsEDgxhcqsYYY1Xs5ogLx<wbr>WPA_2Wlc4&r=BAP9D2FFNY-<wbr>E7nYzFI5fHw&m=<wbr>sdB5SWDdjesA3HFY6oMVN57SUQ3e1b<wbr>D-NFPtNrbSvys&s=<wbr>y0CInmYX6aVEEeJB8dgOXO9lBv0UNk<wbr>6v33OORxrQTLo&e=</a><br>
<br>
<br>
------------------------------<br>
<br>
End of Ipopt Digest, Vol 145, Issue 6<br>
******************************<wbr>*******<br>
</blockquote></div><br></div>