<html>
<head>
<style>
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
</style>
</head>
<body class='hmmessage'>
Hi,<br><br>If that is really the end of the log file, you are indeed having the problem that <br>certain replicas are waiting for communication during exchange, while other<br>have stopped due to reaching -maxh.<br>This problem will also be addressed in 4.1, but for the moment you will have<br>to limit nsteps (as I mailed before) instead of using -maxh.<br><br>PS: if you are interested in the conformations are higher temperatures<br>you should use NPT.<br><br>Berk<br><br><hr id="stopSpelling">From: massimiliano.bonomi@gmail.com<br>To: gmx-users@gromacs.org<br>Subject: Re: [gmx-users] replica restart from checkpoints<br>Date: Fri, 20 Feb 2009 10:47:21 +0100<br><br><br><div><div>On Feb 20, 2009, at 10:07 AM, Berk Hess wrote:</div><br class="EC_Apple-interchange-newline"><blockquote><span class="EC_Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px;"><div class="EC_hmmessage" style="font-size: 10pt; font-family: Verdana;">Hi,<br><br>I guess that actually the -maxh procedure might be the problem in your case.<br>If all replicas stop correctly after -maxh, they will all be between the same exchange events,<br>so it should work.<br>The only issue I can see is that one (or more) replica reaches an exchange attempt step<br>early and waits for communication, while the others are late and get stopped by -maxh.<br>Have you checked that the simulation terminated properly?</div></span></blockquote><div><br></div><div>This is the last output line of one md.log</div><div><br></div><div><div>Step 4834163: Run time exceeded 23.760 hours, will terminate the run</div><div> Step Time Lambda</div><div> 4834164 9668.32800 0.00000</div><div><br></div><div>No checkpoints are created after this point.</div><div>The same for all the other replicas.</div><div>Is this a correct stop or the code should have print out a "final" checkpoint before stopping?</div></div><div><br></div><div>PS: simulations are in the NVT ensemble...</div><div><br></div><div>Massimiliano</div><br><blockquote><span class="EC_Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px;"><div class="EC_hmmessage" style="font-size: 10pt; font-family: Verdana;"><br><br>If this is the case, currently the only solution is not to use -maxh,<br>but to make tpr files with nsteps short enough to finish in time and then use tpbconv<br>to extend the tpr files (without trajectory and energy) and then run mdrun -cpi.<br><br>Berk<br><br><hr id="EC_stopSpelling">From:<span class="EC_Apple-converted-space"> </span><a href="mailto:massimiliano.bonomi@gmail.com">massimiliano.bonomi@gmail.com</a><br>To:<span class="EC_Apple-converted-space"> </span><a href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br>Subject: Re: [gmx-users] replica restart from checkpoints<br>Date: Thu, 19 Feb 2009 22:47:23 +0100<br><br>Thanks for your reply...<div><br><div><blockquote><span class="EC_EC_Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px;"><div class="EC_EC_hmmessage" style="font-size: 10pt; font-family: Verdana;">Which version are you using?<br>In 4.0.3 I made things slightly better by allowing checkpoints<br>to have different step numbers, as long as they fall within<br>the same exchange attempt steps.<br></div></span></blockquote><br><div>I'm using 4.0.3. Same problem with the former versions 4.0.x.</div><br><blockquote><span class="EC_EC_Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px;"><div class="EC_EC_hmmessage" style="font-size: 10pt; font-family: Verdana;">This could still cause problems when the steps in the checkpoints<br>differ very much. But if you use -maxh all simulations should finish<br>closely within each other.</div></span></blockquote><div><br></div>Actually I'm using -maxh!</div><div><br><blockquote><span class="EC_EC_Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px;"><div class="EC_EC_hmmessage" style="font-size: 10pt; font-family: Verdana;"><br>(you can always go back to using tpbconv)<br><br></div></span></blockquote><div><br></div><div>Unfortunately I have no trr files, but just xtc with only solute...</div><br><blockquote><span class="EC_EC_Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px;"><div class="EC_EC_hmmessage" style="font-size: 10pt; font-family: Verdana;">Synchronizing the checkpoint writing is a bit complicated<br>and will probably only be done in 4.1.<br><br></div></span></blockquote><div>Is it not possible to define the writing stride in terms of MD steps?</div><div><br></div><div>Thanks again,</div><div>Massimiliano</div><br><blockquote><span class="EC_EC_Apple-style-span" style="border-collapse: separate; color: rgb(0, 0, 0); font-family: Helvetica; font-size: 12px; font-style: normal; font-variant: normal; font-weight: normal; letter-spacing: normal; line-height: normal; text-indent: 0px; text-transform: none; white-space: normal; word-spacing: 0px;"><div class="EC_EC_hmmessage" style="font-size: 10pt; font-family: Verdana;">Berk<br><br>> From:<span class="EC_EC_Apple-converted-space"> </span><a href="mailto:massimiliano.bonomi@gmail.com">massimiliano.bonomi@gmail.com</a><br>> To:<span class="EC_EC_Apple-converted-space"> </span><a href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br>> Date: Thu, 19 Feb 2009 20:14:15 +0100<br>> Subject: [gmx-users] replica restart from checkpoints<br>><span class="EC_EC_Apple-converted-space"> </span><br>> Dear Gromacs Users,<br>><span class="EC_EC_Apple-converted-space"> </span><br>> I'm experiencing some problems when restarting a replica exchange run<span class="EC_EC_Apple-converted-space"> </span><br>> from previous checkpoint files.<br>> It often happens to me that the number of MD steps done in the<span class="EC_EC_Apple-converted-space"> </span><br>> previous run is not the<br>> same for all the replica. If this is the case, the program stops.<br>> This may happen since checkpoints are written with a stride expressed<br>> in REAL time (every 15 minutes) and replica on different processors<span class="EC_EC_Apple-converted-space"> </span><br>> may have run<br>> for different number of steps in the same amount of time.<br>><span class="EC_EC_Apple-converted-space"> </span><br>> Is it possible to specify the checkpoint writing stride in number of<span class="EC_EC_Apple-converted-space"> </span><br>> steps instead of real time?<br>><span class="EC_EC_Apple-converted-space"> </span><br>> Regards,<br>> Massimiliano Bonomi<br>> _______________________________________________<br>> gmx-users mailing list<span class="EC_EC_Apple-converted-space"> </span><a href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br>><span class="EC_EC_Apple-converted-space"> </span><a href="http://www.gromacs.org/mailman/listinfo/gmx-users">http://www.gromacs.org/mailman/listinfo/gmx-users</a><br>> Please search the archive at<span class="EC_EC_Apple-converted-space"> </span><a href="http://www.gromacs.org/search">http://www.gromacs.org/search</a><span class="EC_EC_Apple-converted-space"> </span>before posting!<br>> Please don't post (un)subscribe requests to the list. Use the<span class="EC_EC_Apple-converted-space"> </span><br>> www interface or send it to<span class="EC_EC_Apple-converted-space"> </span><a href="mailto:gmx-users-request@gromacs.org">gmx-users-request@gromacs.org</a>.<br>> Can't post? Read<span class="EC_EC_Apple-converted-space"> </span><a href="http://www.gromacs.org/mailing_lists/users.php">http://www.gromacs.org/mailing_lists/users.php</a><br><br><hr>Express yourself instantly with MSN Messenger!<span class="EC_EC_Apple-converted-space"> </span><a href="http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/">MSN Messenger</a><span class="EC_EC_Apple-converted-space"> </span>_______________________________________________<br>gmx-users mailing list <a href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br><a href="http://www.gromacs.org/mailman/listinfo/gmx-users">http://www.gromacs.org/mailman/listinfo/gmx-users</a><br>Please search the archive at<span class="EC_EC_Apple-converted-space"> </span><a href="http://www.gromacs.org/search">http://www.gromacs.org/search</a><span class="EC_EC_Apple-converted-space"> </span>before posting!<br>Please don't post (un)subscribe requests to the list. Use the<span class="EC_EC_Apple-converted-space"> </span><br>www interface or send it to<span class="EC_EC_Apple-converted-space"> </span><a href="mailto:gmx-users-request@gromacs.org">gmx-users-request@gromacs.org</a>.<br>Can't post? Read<span class="EC_EC_Apple-converted-space"> </span><a href="http://www.gromacs.org/mailing_lists/users.php">http://www.gromacs.org/mailing_lists/users.php</a></div></span></blockquote></div><br></div><br><hr>What can you do with the new Windows Live?<span class="EC_Apple-converted-space"> </span><a href="http://www.microsoft.com/windows/windowslive/default.aspx">Find out</a><span class="EC_Apple-converted-space"> </span>_______________________________________________<br>gmx-users mailing list <a href="mailto:gmx-users@gromacs.org">gmx-users@gromacs.org</a><br><a href="http://www.gromacs.org/mailman/listinfo/gmx-users">http://www.gromacs.org/mailman/listinfo/gmx-users</a><br>Please search the archive at<span class="EC_Apple-converted-space"> </span><a href="http://www.gromacs.org/search">http://www.gromacs.org/search</a><span class="EC_Apple-converted-space"> </span>before posting!<br>Please don't post (un)subscribe requests to the list. Use the<span class="EC_Apple-converted-space"> </span><br>www interface or send it to<span class="EC_Apple-converted-space"> </span><a href="mailto:gmx-users-request@gromacs.org">gmx-users-request@gromacs.org</a>.<br>Can't post? Read<span class="EC_Apple-converted-space"> </span><a href="http://www.gromacs.org/mailing_lists/users.php">http://www.gromacs.org/mailing_lists/users.php</a></div></span></blockquote></div><br><br /><hr />Express yourself instantly with MSN Messenger! <a href='http://clk.atdmt.com/AVE/go/onm00200471ave/direct/01/' target='_new'>MSN Messenger</a></body>
</html>