Should the group of my user be None?
Hi all, I just installed cygwin on my Windows 8.1 laptop and I found that the result of ls -l is like this: -rw-rw-r-- 1 Theodore None 0 Nov 8 22:44 a And I fond that I am in several groups $ groups Theodore Theodore : None root Performance Log Users This raise my curiosity because when I use git, I got some error. Then I read this: http://stackoverflow.com/questions/9561759/why-cannot-chmod-in-cygwin-on-windows-8-cp I wonder why I need to use chgrp to make it right? Is this a bug? Thanks Theo -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Should the group of my user be None?
Shouldn't I be in the group with the same name of my username, like in Linux? 在 11/8/2014 11:29 PM, Theodore Si 写道: Hi all, I just installed cygwin on my Windows 8.1 laptop and I found that the result of ls -l is like this: -rw-rw-r-- 1 Theodore None 0 Nov 8 22:44 a And I fond that I am in several groups $ groups Theodore Theodore : None root Performance Log Users This raise my curiosity because when I use git, I got some error. Then I read this: http://stackoverflow.com/questions/9561759/why-cannot-chmod-in-cygwin-on-windows-8-cp I wonder why I need to use chgrp to make it right? Is this a bug? Thanks Theo -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Should the group of my user be None?
Thank you for your replies. The permission of files under ~/.ssh can't be changed to 600 when their group owner is None. I have to chgrp -R Users (or Administrators, or any other group name other than None) ~/.ssh to make it possible to run chmod on them. I suppose this is a bug of cygwin on Windows 8/8.1 ? 在 11/9/2014 11:05 AM, Larry Hall (Cygwin) 写道: On 11/08/2014 11:17 AM, Theodore Si wrote: Shouldn't I be in the group with the same name of my username, like in Linux? No. Windows isn't Linux. Of course, if you want to make a group with your user name and add your user to that group, Windows will probably let you do that. But that's not a convention for user accounts on Windows. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: Should the group of my user be None?
Thank you for your replies. The permission of files under ~/.ssh can't be changed to 600 when their group owner is None. I have to chgrp -R Users (or Administrators, or any other group name other than None) ~/.ssh to make it possible to run chmod on them. I suppose this is a bug of cygwin on Windows 8/8.1 ? I changed my GID in /etc/passwd from 513 to 544 to make my primary group Administrators. Now the group owner of the files are turned to ? . Is it OK to do this? 在 11/9/2014 11:33 AM, Theodore Si 写道: Thank you for your replies. The permission of files under ~/.ssh can't be changed to 600 when their group owner is None. I have to chgrp -R Users (or Administrators, or any other group name other than None) ~/.ssh to make it possible to run chmod on them. I suppose this is a bug of cygwin on Windows 8/8.1 ? 在 11/9/2014 11:05 AM, Larry Hall (Cygwin) 写道: On 11/08/2014 11:17 AM, Theodore Si wrote: Shouldn't I be in the group with the same name of my username, like in Linux? No. Windows isn't Linux. Of course, if you want to make a group with your user name and add your user to that group, Windows will probably let you do that. But that's not a convention for user accounts on Windows. -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Re: failure notice
Thank you for your replies. The permission of files under ~/.ssh can't be changed to 600 when their group owner is None. I have to chgrp -R Users (or Administrators, or any other group name other than None) ~/.ssh to make it possible to run chmod on them. I suppose this is a bug of cygwin on Windows 8/8.1 ? I changed my GID in /etc/passwd from 513 to 544 to make my primary group Administrators. Now the group owner of the files are turned to ? . Is it OK to do this? On Sun, Nov 9, 2014 at 11:40 AM, mailer-dae...@sourceware.org wrote: Hi. This is the qmail-send program at sourceware.org. I'm afraid I wasn't able to deliver your message to the following addresses. This is a permanent error; I've given up. Sorry it didn't work out. cygwin@cygwin.com: Invalid mime type text/html detected in message text or attachment. Please send plain text messages only. See http://sourceware.org/lists.html#sourceware-list-info for more information. Contact cygwin-ow...@cygwin.com if you have questions about this. (#5.7.2) --- Below this line is a copy of the message. Return-Path: sjyz...@gmail.com Received: (qmail 13527 invoked by uid 89); 9 Nov 2014 03:40:17 - Authentication-Results: sourceware.org; auth=none X-Virus-Checked: by ClamAV 0.98.4 on sourceware.org X-Virus-Found: No X-Spam-SWARE-Status: No, score=3.8 required=5.0 tests=BAYES_05,FREEMAIL_FROM,HTML_MESSAGE,RCVD_IN_DNSWL_LOW,SPF_PASS autolearn=no version=3.3.2 X-Spam-Status: No, score=3.8 required=5.0 tests=BAYES_05,FREEMAIL_FROM,HTML_MESSAGE,RCVD_IN_DNSWL_LOW,SPF_PASS autolearn=no version=3.3.2 X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on sourceware.org X-Spam-Level: *** X-HELO: mail-oi0-f50.google.com Received: from mail-oi0-f50.google.com (HELO mail-oi0-f50.google.com) (209.85.218.50) by sourceware.org (qpsmtpd/0.93/v0.84-503-g423c35a) with (AES128-SHA encrypted) ESMTPS; Sun, 09 Nov 2014 03:40:15 + Received: by mail-oi0-f50.google.com with SMTP id v63so4113037oia.9 for cygwin@cygwin.com; Sat, 08 Nov 2014 19:40:13 -0800 (PST) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:from:date:message-id:subject:to :content-type; bh=gLcsDdYrlI6xfgUXkVOcs/PuOHK4MCC7vay05Eu2bpE=; b=lVxGzQBup/DBjWJqJ4n1j4KiBPx4bCQYHTFo3oyOJAH4wMH76Oq0tyZL7trRtrbel9 1SxN20Ody2bVMvKZD98LLL6wjiCevxs/LmmUjVPwec0FbRoA/Rx5YSMo5UHacOePTjwh SCC0vHBT9Wi7iiwuZIXvdWs0NYvkuBckOEEI4z8wYpbdvEBs2cmXf7L+QhmcbBXSkLVg n93rSczppuEmVTq0y8WzfTFLro2/NZ/ikToJQ3gNol+EwOhUahvjdMubZ5YHdM0RGvTJ DLZItS4U1gZshyODCtCrNoUNGX7XjrhYKNvxiywwCp7lcase660ufoKb9ugeb7MrExqV fS2Q== X-Received: by 10.182.32.33 with SMTP id f1mr18506071obi.34.1415504413720; Sat, 08 Nov 2014 19:40:13 -0800 (PST) MIME-Version: 1.0 Received: by 10.60.59.197 with HTTP; Sat, 8 Nov 2014 19:39:43 -0800 (PST) In-Reply-To: 545ed9de.5010...@cygwin.com References: 545e36e5.5010...@gmail.com 545e420a.2020...@gmail.com 545ed9de.5010...@cygwin.com From: Theodore Si sjyz...@gmail.com Date: Sun, 9 Nov 2014 11:39:43 +0800 Message-ID: CAG=ehjo-ojefqnmysungewpkq1kpjnnaebs9axgkrcfuozk...@mail.gmail.com Subject: Re: Should the group of my user be None? To: cygwin@cygwin.com Content-Type: multipart/alternative; boundary=089e013a0796c206b7050764cd18 --089e013a0796c206b7050764cd18 Content-Type: text/plain; charset=UTF-8 Thank you for your replies. The permission of files under ~/.ssh can't be changed to 600 when their group owner is None. I have to chgrp -R Users (or Administrators, or any other group name other than None) ~/.ssh to make it possible to run chmod on them. I suppose this is a bug of cygwin on Windows 8/8.1 ? I changed my GID in /etc/passwd from 513 to 544 to make my primary group Administrators. Now the group owner of the files are turned to ? . Is it OK to do this? On Sun, Nov 9, 2014 at 11:05 AM, Larry Hall (Cygwin) reply-to-list-only...@cygwin.com wrote: On 11/08/2014 11:17 AM, Theodore Si wrote: Shouldn't I be in the group with the same name of my username, like in Linux? No. Windows isn't Linux. Of course, if you want to make a group with your user name and add your user to that group, Windows will probably let you do that. But that's not a convention for user accounts on Windows. -- Larry _ A: Yes. Q: Are you sure? A: Because it reverses the logical flow of conversation. Q: Why is top posting annoying in email? -- Problem reports: http://cygwin.com/problems.html FAQ: http://cygwin.com/faq/ Documentation: http://cygwin.com/docs.html Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple --089e013a0796c206b7050764cd18 Content-Type: text/html; charset=UTF-8 Content-Transfer-Encoding: quoted-printable div dir=3DltrThank you
How can I set the IP a worker use?
Hi all, I have two network interface card on one node, one is a Eithernet card, the other Infiniband HCA. The master has two IP addresses, lets say 1.2.3.4 (for Eithernet card) and 2.3.4.5 (for HCA). I can start the master by export SPARK_MASTER_IP='1.2.3.4';sbin/start-master.sh to let master listen on 1.2.3.4:7077 But when I connect the worker to the master by using spark-1.0.1/bin/spark-class org.apache.spark.deploy.worker.Worker spark://1.2.3.4:7077 I will get errors, since it is using its HCA card. How can I let the worker use its Eithernet card? Thanks
Re: How can I set the IP a worker use?
I found this. So it seems that we should use -h or --host instead of -i and --ip. -i HOST, --ip IP Hostname to listen on (deprecated, please use --host or -h) -h HOST, --host HOST Hostname to listen on 在 10/24/2014 3:35 PM, Akhil Das 写道: Try using the --ip parameter http://spark.apache.org/docs/latest/spark-standalone.html#starting-a-cluster-manually while starting the worker. like: spark-1.0.1/bin/spark-class org.apache.spark.deploy.worker.Worker --ip 1.2.3.4 spark://1.2.3.4:7077 http://1.2.3.4:7077 Thanks Best Regards On Fri, Oct 24, 2014 at 12:34 PM, Theodore Si sjyz...@gmail.com mailto:sjyz...@gmail.com wrote: Hi all, I have two network interface card on one node, one is a Eithernet card, the other Infiniband HCA. The master has two IP addresses, lets say 1.2.3.4 (for Eithernet card) and 2.3.4.5 (for HCA). I can start the master by export SPARK_MASTER_IP='1.2.3.4';sbin/start-master.sh to let master listen on 1.2.3.4:7077 http://1.2.3.4:7077 But when I connect the worker to the master by using spark-1.0.1/bin/spark-class org.apache.spark.deploy.worker.Worker spark://1.2.3.4:7077 http://1.2.3.4:7077 I will get errors, since it is using its HCA card. How can I let the worker use its Eithernet card? Thanks
Which part of the code deals with communication?
Hi all, Workers will exchange data in between, right? What classes are in charge of these actions? - To unsubscribe, e-mail: dev-unsubscr...@spark.apache.org For additional commands, e-mail: dev-h...@spark.apache.org
Re: Initial job has not accepted any resources when launching SparkPi example on a worker.
Can anyone help me, please? 在 10/14/2014 9:58 PM, Theodore Si 写道: Hi all, I have two nodes, one as master(*host1*) and the other as worker(*host2*). I am using the standalone mode. After starting the master on host1, I run $ export MASTER=spark://host1:7077 $ bin/run-example SparkPi 10 on host2, but I get this: 14/10/14 21:54:23 WARN TaskSchedulerImpl: Initial job has not accepted any resources; check your cluster UI to ensure that workers are registered and have sufficient memory And it repeats again and again. How can I fix this? Best Regards Theo
Initial job has not accepted any resources when launching SparkPi example on a worker.
Hi all, I have two nodes, one as master(*host1*) and the other as worker(*host2*). I am using the standalone mode. After starting the master on host1, I run $ export MASTER=spark://host1:7077 $ bin/run-example SparkPi 10 on host2, but I get this: 14/10/14 21:54:23 WARN TaskSchedulerImpl: Initial job has not accepted any resources; check your cluster UI to ensure that workers are registered and have sufficient memory And it repeats again and again. How can I fix this? Best Regards Theo
Can I run examples on cluster?
Hi all, I want to use two nodes for test, one as master, the other worker. Can I submit the example application included in Spark source code tarball on master to let it run on the worker? What should I do? BR, Theo - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
Re: Can I run examples on cluster?
But I cannot do this via using ./bin/run-example SparkPi 10 right? On Fri, Oct 10, 2014 at 6:04 PM, Akhil Das ak...@sigmoidanalytics.com wrote: This is how the spark-cluster looks like So your driver program (example application) can be ran on the master (or anywhere which has access to the master - clustermanager) and the workers will execute it. Thanks Best Regards On Fri, Oct 10, 2014 at 2:47 PM, Theodore Si sjyz...@gmail.com wrote: Hi all, I want to use two nodes for test, one as master, the other worker. Can I submit the example application included in Spark source code tarball on master to let it run on the worker? What should I do? BR, Theo - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
Re: Can I run examples on cluster?
Should I pack the example into a jar file and submit it on master? On Fri, Oct 10, 2014 at 9:32 PM, Theodore Si sjyz...@gmail.com wrote: But I cannot do this via using ./bin/run-example SparkPi 10 right? On Fri, Oct 10, 2014 at 6:04 PM, Akhil Das ak...@sigmoidanalytics.com wrote: This is how the spark-cluster looks like So your driver program (example application) can be ran on the master (or anywhere which has access to the master - clustermanager) and the workers will execute it. Thanks Best Regards On Fri, Oct 10, 2014 at 2:47 PM, Theodore Si sjyz...@gmail.com wrote: Hi all, I want to use two nodes for test, one as master, the other worker. Can I submit the example application included in Spark source code tarball on master to let it run on the worker? What should I do? BR, Theo - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
What if I port Spark from TCP/IP to RDMA?
Hi, Let's say that I managed to port Spark from TCP/IP to RDMA. What tool or benchmark can I use to test the performance improvement? BR, Theo - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
How to benchmark SPARK apps?
Hi all, What tools should I use to benchmark SPARK applications? BR, Theo - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
Re: How to benchmark SPARK applications?
How can I get figures like those in the Evaluation part of the following paper? http://www.cs.berkeley.edu/~matei/papers/2011/tr_spark.pdf 在 10/10/2014 10:35 AM, Theodore Si 写道: Hi all, What tools should I use to benchmark SPARK applications? BR, Theo - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
Re: How to benchmark SPARK apps?
What can I get from it? Can you show me some results please? 在 10/10/2014 10:46 AM, 牛兆捷 写道: *You can try https://github.com/databricks/spark-perf* - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
Re: [gmx-users] which item is gpu wait for cpu?
That's a breakdown of the timings that CUDA cannot always measure. What does that mean? We run the same test case on the same machine. This GPU timings part didn't show up before, but now it appears. 在 9/25/2014 6:11 PM, Mark Abraham 写道: On Thu, Sep 25, 2014 at 11:57 AM, xiexiao...@sjtu.edu.cn xiexiao...@sjtu.edu.cn wrote: Thanks.And I have another question want to ask. I used the command mdrun_mpi -v -deffnm test to run the .tpr file and find some new information in my log file: GPU timings - Computing: Count Wall t (s) ms/step % - Pair list H2D 501 0.364 0.726 0.4 X / q H2D 20001 4.153 0.208 4.5 Nonbonded F kernel 18800 77.657 4.131 84.5 Nonbonded F+ene k. 700 4.060 5.800 4.4 Nonbonded F+prune k. 400 2.121 5.303 2.3 Nonbonded F+ene+prune k. 101 0.717 7.103 0.8 F D2H 20001 2.873 0.144 3.1 - Total 91.946 4.597 100.0 - Force evaluation time GPU/CPU: 4.597 ms/5.245 ms = 0.876. But I used to use the some command and didn't appear such information.Does anyone know why? That's a breakdown of the timings that CUDA cannot always measure. I also find that the time consuming of Nonbonded F kernel is 77.657s but the time consuming of force is only 55.263s.Does the force item include non-bonded force? No, it's non-PME work being done on the CPU (e.g. bonds, restraints, free-energy short-ranged) Mark If it was true,why the Nonbonded F kernel takes more time than the force ? xiexiao...@sjtu.edu.cn From: Szilárd Páll Date: 2014-09-24 17:45 To: Discussion list for GROMACS users CC: gromacs.org_gmx-users Subject: Re: [gmx-users] which item is gpu wait for cpu? On Wed, Sep 24, 2014 at 4:57 AM, xiexiao...@sjtu.edu.cn xiexiao...@sjtu.edu.cn wrote: I think that the items (wait GPU nonlocal) and (wait GPU local) in log file are the time consuming that cpu wait for gpu. And Does anyone know that which item is the time consuming that gpu wait for cpu in log file? Due to a CUDA runtime limitation, that can not measured except for the no domain-decomposition case (where it is expressed as overlap). xiexiao...@sjtu.edu.cn -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Where can I find the module diagram of SPARK?
Hi, Please help me with that. BR, Theodore Si - To unsubscribe, e-mail: user-unsubscr...@spark.apache.org For additional commands, e-mail: user-h...@spark.apache.org
[gmx-users] How are the clusters of particles created?
Hi all, In the page 34 of the manual: The Verlet cut-off scheme is implemented in a very efficient fashion based on clusters of particles. The simplest example is a cluster size of 4 particles. The pair list is then constructed based on cluster pairs. I want to know on what condition will 4 particles be put in a cluster? Can you explain the algorithm in a simple manner? BR, Theo -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] On what scale will simulation with PME-dedicated nodes perform better?
Hi all, I run gromacs 4.6 on 5 nodes(each has 16 CPU cores and 2 Nvidia K20m) and 4 nodes in the following ways: 5 nodes: 1. Each node has 8 MPI processes, and use one node as PME-dedicated node 2. Each node has 8 MPI processes, and use two nodes as PME-dedicated nodes 3. Each node has 4 MPI processes, and use one node as PME-dedicated node In these settings, the log files complain that PME nodes have more work to do than PP nodes, and the average imbalance is 20% - 40%. 4nodes: Each node has 8 MPI processes, and there is no PME-dedicated node In the log file, the PME mesh wall time is about the half compared the settings above. My guess is that the scaling of my run is small so PME-dedicated nodes won't do any good. So, on what condition should I set PME nodes manually? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Can we set the number of pure PME nodes when using GPUCPU?
I mapped 2 GPUs to multiple MPI ranks by using -gpu_id On 8/26/2014 1:12 AM, Xingcheng Lin wrote: Theodore Si sjyzhxw@... writes: Hi, https://onedrive.live.com/redir? resid=990FCE59E48164A4!2572authkey=!AP82sTNxS6MHgUkithint=file%2clog https://onedrive.live.com/redir? resid=990FCE59E48164A4!2482authkey=!APLkizOBzXtPHxsithint=file%2clog These are 2 log files. The first one is using 64 cpu cores(64 / 16 = 4 nodes) and 4nodes*2 = 8 GPUs, and the second is using 512 cpu cores, no GPU. When we look at the 64 cores log file, we find that in the R E A L C Y C L E A N D T I M E A C C O U N T I N G table, the total wall time is the sum of every line, that is 37.730=2.201+0.082+...+1.150. So we think that when the CPUs is doing PME, GPUs are doing nothing. That's why we say they are working sequentially. As for the 512 cores log file, the total wall time is approximately the sum of PME mesh and PME wait for PP. We think this is because PME-dedicated nodes finished early, and the total wall time is the time spent on PP nodes, therefore time spent on PME is covered. Hi, I have a naive question: In your log file there are only 2 GPUs being detected: 2 GPUs detected on host gpu42: #0: NVIDIA Tesla K20m, compute cap.: 3.5, ECC: yes, stat: compatible #1: NVIDIA Tesla K20m, compute cap.: 3.5, ECC: yes, stat: compatible In the end you selected 8 GPUs 8 GPUs user-selected for this run: #0, #0, #0, #0, #1, #1, #1, #1 Did you choose 8 GPUs or 2 GPUs? What is your mdrun command? Thank you, -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Can we set the number of pure PME nodes when using GPUCPU?
Hi Szilárd, But CUDA 5.5 won't work with icc 14, right? It only works with 12.1 unless a header of CUDA 5.5 to be modified. Theo On 8/25/2014 9:44 PM, Szilárd Páll wrote: On Mon, Aug 25, 2014 at 8:08 AM, Mark Abraham mark.j.abra...@gmail.com wrote: On Mon, Aug 25, 2014 at 5:01 AM, Theodore Si sjyz...@gmail.com wrote: Hi, https://onedrive.live.com/redir?resid=990FCE59E48164A4! 2572authkey=!AP82sTNxS6MHgUkithint=file%2clog https://onedrive.live.com/redir?resid=990FCE59E48164A4! 2482authkey=!APLkizOBzXtPHxsithint=file%2clog These are 2 log files. The first one is using 64 cpu cores(64 / 16 = 4 nodes) and 4nodes*2 = 8 GPUs, and the second is using 512 cpu cores, no GPU. When we look at the 64 cores log file, we find that in the R E A L C Y C L E A N D T I M E A C C O U N T I N G table, the total wall time is the sum of every line, that is 37.730=2.201+0.082+...+1.150. So we think that when the CPUs is doing PME, GPUs are doing nothing. That's why we say they are working sequentially. Please note that sequential means one phase after another. Your log files don't show the timing breakdown for the GPUs, which is distinct from showing that the GPUs ran and then the CPUs ran (which I don't think the code even permits!). References to CUDA 8x8 kernels do show the GPU was active. There was an issue with mdrun not always being able to gather and publish the GPU timing results; I don't recall the conditions (Szilard might remember), but it might be fixed in a later release. It is a limitation (well, I'd say borderline bug) in CUDA that if you have multiple work-queues (=streams), reliable timing using the CUDA built-in mechanisms is impossible. There may be a way to work around this, but that won't happen in the current versions. What's important is to observe the wait time on the CPU sideand of course, if the Op is profiling this is not an issue. In any case, you should probably be doing performance optimization on a GROMACS version that isn't a year old. I gather that you didn't actually observe the GPUs idle - e.g. with a performance monitoring tool? Otherwise, and in the absence of a description of your simulation system, I'd say that log file looks somewhere between normal and optimal. For the record, for better performance, you should probably be following the advice of the install guide and not compiling FFTW with AVX support, and using one of the five gcc minor versions released since 4.4 ;-) And besides avoiding ancient gcc versions, I suggest using CUDA 5.5 (which you can use because you have version 5.5 driver which I see in your log file): Additionally, I suggest avoiding MKL and using FFTW instead. For the grid sizes of our interest all benchmarks I did in the past showed considerably higher FFTW performance. Same goes for icc, but feel free to benchmark and please report back if you find the opposite. As for the 512 cores log file, the total wall time is approximately the sum of PME mesh and PME wait for PP. We think this is because PME-dedicated nodes finished early, and the total wall time is the time spent on PP nodes, therefore time spent on PME is covered. Yes, using an offload model makes it awkward to report CPU timings, because there are two kinds of CPU ranks. The total of the Wall t column adds up to twice the total time taken (which is noted explicitly in more recent mdrun versions). By design, the PME ranks do finish early, as you know from Figure 3.16 of the manual. As you can see in the table, the PP ranks spend 26% of their time waiting for the results from the PME ranks, and this is the origin of the note (above the table) that you might want to balance things better. Mark On 8/23/2014 9:30 PM, Mark Abraham wrote: On Sat, Aug 23, 2014 at 1:47 PM, Theodore Si sjyz...@gmail.com wrote: Hi, When we used 2 GPU nodes (each has 2 cpus and 2 gpus) to do a mdrun(with no PME-dedicated node), we noticed that when CPU are doing PME, GPU are idle, That could happen if the GPU completes its work too fast, in which case the end of the log file will probably scream about imbalance. that is they are doing their work sequentially. Highly unlikely, not least because the code is written to overlap the short-range work on the GPU with everything else on the CPU. What's your evidence for *sequential* rather than *imbalanced*? Is it supposed to be so? No, but without seeing your .log files, mdrun command lines and knowing about your hardware, there's nothing we can say. Is it the same reason as GPUs on PME-dedicated nodes won't be used during a run like you said before? Why would you suppose that? I said GPUs do work from the PP ranks on their node. That's true here. So if we want to exploit our hardware, we should map PP-PME ranks manually, right? Say, use one node as PME-dedicated node and leave the GPUs on that node idle, and use two nodes to do the other stuff. How do you think about this arrangement? Probably a terrible idea. You should
Re: [gmx-users] Can we set the number of pure PME nodes when using GPUCPU?
Hi, https://onedrive.live.com/redir?resid=990FCE59E48164A4!2572authkey=!AP82sTNxS6MHgUkithint=file%2clog https://onedrive.live.com/redir?resid=990FCE59E48164A4!2482authkey=!APLkizOBzXtPHxsithint=file%2clog These are 2 log files. The first one is using 64 cpu cores(64 / 16 = 4 nodes) and 4nodes*2 = 8 GPUs, and the second is using 512 cpu cores, no GPU. When we look at the 64 cores log file, we find that in the R E A L C Y C L E A N D T I M E A C C O U N T I N G table, the total wall time is the sum of every line, that is 37.730=2.201+0.082+...+1.150. So we think that when the CPUs is doing PME, GPUs are doing nothing. That's why we say they are working sequentially. As for the 512 cores log file, the total wall time is approximately the sum of PME mesh and PME wait for PP. We think this is because PME-dedicated nodes finished early, and the total wall time is the time spent on PP nodes, therefore time spent on PME is covered. On 8/23/2014 9:30 PM, Mark Abraham wrote: On Sat, Aug 23, 2014 at 1:47 PM, Theodore Si sjyz...@gmail.com wrote: Hi, When we used 2 GPU nodes (each has 2 cpus and 2 gpus) to do a mdrun(with no PME-dedicated node), we noticed that when CPU are doing PME, GPU are idle, That could happen if the GPU completes its work too fast, in which case the end of the log file will probably scream about imbalance. that is they are doing their work sequentially. Highly unlikely, not least because the code is written to overlap the short-range work on the GPU with everything else on the CPU. What's your evidence for *sequential* rather than *imbalanced*? Is it supposed to be so? No, but without seeing your .log files, mdrun command lines and knowing about your hardware, there's nothing we can say. Is it the same reason as GPUs on PME-dedicated nodes won't be used during a run like you said before? Why would you suppose that? I said GPUs do work from the PP ranks on their node. That's true here. So if we want to exploit our hardware, we should map PP-PME ranks manually, right? Say, use one node as PME-dedicated node and leave the GPUs on that node idle, and use two nodes to do the other stuff. How do you think about this arrangement? Probably a terrible idea. You should identify the cause of the imbalance, and fix that. Mark Theo On 8/22/2014 7:20 PM, Mark Abraham wrote: Hi, Because no work will be sent to them. The GPU implementation can accelerate domains from PP ranks on their node, but with an MPMD setup that uses dedicated PME nodes, there will be no PP ranks on nodes that have been set up with only PME ranks. The two offload models (PP work - GPU; PME work - CPU subset) do not work well together, as I said. One can devise various schemes in 4.6/5.0 that could use those GPUs, but they either require * each node does both PME and PP work (thus limiting scaling because of the all-to-all for PME, and perhaps making poor use of locality on multi-socket nodes), or * that all nodes have PP ranks, but only some have PME ranks, and the nodes map their GPUs to PP ranks in a way that is different depending on whether PME ranks are present (which could work well, but relies on the DD load-balancer recognizing and taking advantage of the faster progress of the PP ranks that have better GPU support, and requires that you get very dirty hands laying out PP and PME ranks onto hardware that will later match the requirements of the DD load balancer, and probably that you balance PP-PME load manually) I do not recommend the last approach, because of its complexity. Clearly there are design decisions to improve. Work is underway. Cheers, Mark On Fri, Aug 22, 2014 at 10:11 AM, Theodore Si sjyz...@gmail.com wrote: Hi Mark, Could you tell me why that when we are GPU-CPU nodes as PME-dedicated nodes, the GPU on such nodes will be idle? Theo On 8/11/2014 9:36 PM, Mark Abraham wrote: Hi, What Carsten said, if running on nodes that have GPUs. If running on a mixed setup (some nodes with GPU, some not), then arranging your MPI environment to place PME ranks on CPU-only nodes is probably worthwhile. For example, all your PP ranks first, mapped to GPU nodes, then all your PME ranks, mapped to CPU-only nodes, and then use mdrun -ddorder pp_pme. Mark On Mon, Aug 11, 2014 at 2:45 AM, Theodore Si sjyz...@gmail.com wrote: Hi Mark, This is information of our cluster, could you give us some advice as regards to our cluster so that we can make GMX run faster on our system? Each CPU node has 2 CPUs and each GPU node has 2 CPUs and 2 Nvidia K20M Device Name Device Type Specifications Number CPU NodeIntelH2216JFFKRNodesCPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 332 Fat NodeIntelH2216WPFKRNodesCPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 256G(16×16G) ECC Registered DDR3 1600MHz Samsung Memory20 GPU Node
Re: [gmx-users] Can we set the number of pure PME nodes when using GPUCPU?
Hi, When we used 2 GPU nodes (each has 2 cpus and 2 gpus) to do a mdrun(with no PME-dedicated node), we noticed that when CPU are doing PME, GPU are idle, that is they are doing their work sequentially. Is it supposed to be so? Is it the same reason as GPUs on PME-dedicated nodes won't be used during a run like you said before? So if we want to exploit our hardware, we should map PP-PME ranks manually, right? Say, use one node as PME-dedicated node and leave the GPUs on that node idle, and use two nodes to do the other stuff. How do you think about this arrangement? Theo On 8/22/2014 7:20 PM, Mark Abraham wrote: Hi, Because no work will be sent to them. The GPU implementation can accelerate domains from PP ranks on their node, but with an MPMD setup that uses dedicated PME nodes, there will be no PP ranks on nodes that have been set up with only PME ranks. The two offload models (PP work - GPU; PME work - CPU subset) do not work well together, as I said. One can devise various schemes in 4.6/5.0 that could use those GPUs, but they either require * each node does both PME and PP work (thus limiting scaling because of the all-to-all for PME, and perhaps making poor use of locality on multi-socket nodes), or * that all nodes have PP ranks, but only some have PME ranks, and the nodes map their GPUs to PP ranks in a way that is different depending on whether PME ranks are present (which could work well, but relies on the DD load-balancer recognizing and taking advantage of the faster progress of the PP ranks that have better GPU support, and requires that you get very dirty hands laying out PP and PME ranks onto hardware that will later match the requirements of the DD load balancer, and probably that you balance PP-PME load manually) I do not recommend the last approach, because of its complexity. Clearly there are design decisions to improve. Work is underway. Cheers, Mark On Fri, Aug 22, 2014 at 10:11 AM, Theodore Si sjyz...@gmail.com wrote: Hi Mark, Could you tell me why that when we are GPU-CPU nodes as PME-dedicated nodes, the GPU on such nodes will be idle? Theo On 8/11/2014 9:36 PM, Mark Abraham wrote: Hi, What Carsten said, if running on nodes that have GPUs. If running on a mixed setup (some nodes with GPU, some not), then arranging your MPI environment to place PME ranks on CPU-only nodes is probably worthwhile. For example, all your PP ranks first, mapped to GPU nodes, then all your PME ranks, mapped to CPU-only nodes, and then use mdrun -ddorder pp_pme. Mark On Mon, Aug 11, 2014 at 2:45 AM, Theodore Si sjyz...@gmail.com wrote: Hi Mark, This is information of our cluster, could you give us some advice as regards to our cluster so that we can make GMX run faster on our system? Each CPU node has 2 CPUs and each GPU node has 2 CPUs and 2 Nvidia K20M Device Name Device Type Specifications Number CPU NodeIntelH2216JFFKRNodesCPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 332 Fat NodeIntelH2216WPFKRNodesCPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 256G(16×16G) ECC Registered DDR3 1600MHz Samsung Memory20 GPU NodeIntelR2208GZ4GC CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 50 MIC NodeIntelR2208GZ4GC CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 5 Computing Network SwitchMellanox Infiniband FDR Core Switch 648× FDR Core Switch MSX6536-10R, Mellanox Unified Fabric Manager 1 Mellanox SX1036 40Gb Switch 36× 40Gb Ethernet Switch SX1036, 36× QSFP Interface 1 Management Network Switch Extreme Summit X440-48t-10G 2-layer Switch 48× 1Giga Switch Summit X440-48t-10G, authorized by ExtremeXOS 9 Extreme Summit X650-24X 3-layer Switch 24× 10Giga 3-layer Ethernet Switch Summit X650-24X, authorized by ExtremeXOS1 Parallel StorageDDN Parallel Storage System DDN SFA12K Storage System 1 GPU GPU Accelerator NVIDIA Tesla Kepler K20M70 MIC MIC Intel Xeon Phi 5110P Knights Corner 10 40Gb Ethernet Card MCX314A-BCBTMellanox ConnextX-3 Chip 40Gb Ethernet Card 2× 40Gb Ethernet ports, enough QSFP cables 16 SSD Intel SSD910Intel SSD910 Disk, 400GB, PCIE 80 On 8/10/2014 5:50 AM, Mark Abraham wrote: That's not what I said You can set... -npme behaves the same whether or not GPUs are in use. Using separate ranks for PME caters to trying to minimize the cost of the all-to-all communication of the 3DFFT. That's still relevant when using GPUs, but if separate PME ranks are used, any GPUs on nodes that only have PME ranks are left idle. The most effective approach depends critically on the hardware and simulation setup, and whether you pay money
[gmx-users] Why we are not using GPU to solve FFT?
Hi, I wonder why we are using cpu instead of gpu to solve FFT? Is it possible to use gpu fft library, say cuFFT to make the FFT used in PME faster? BR, Theo -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Can we set the number of pure PME nodes when using GPUCPU?
Hi Mark, Could you tell me why that when we are GPU-CPU nodes as PME-dedicated nodes, the GPU on such nodes will be idle? Theo On 8/11/2014 9:36 PM, Mark Abraham wrote: Hi, What Carsten said, if running on nodes that have GPUs. If running on a mixed setup (some nodes with GPU, some not), then arranging your MPI environment to place PME ranks on CPU-only nodes is probably worthwhile. For example, all your PP ranks first, mapped to GPU nodes, then all your PME ranks, mapped to CPU-only nodes, and then use mdrun -ddorder pp_pme. Mark On Mon, Aug 11, 2014 at 2:45 AM, Theodore Si sjyz...@gmail.com wrote: Hi Mark, This is information of our cluster, could you give us some advice as regards to our cluster so that we can make GMX run faster on our system? Each CPU node has 2 CPUs and each GPU node has 2 CPUs and 2 Nvidia K20M Device Name Device Type Specifications Number CPU NodeIntelH2216JFFKRNodesCPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 332 Fat NodeIntelH2216WPFKRNodesCPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 256G(16×16G) ECC Registered DDR3 1600MHz Samsung Memory20 GPU NodeIntelR2208GZ4GC CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 50 MIC NodeIntelR2208GZ4GC CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 5 Computing Network SwitchMellanox Infiniband FDR Core Switch 648× FDR Core Switch MSX6536-10R, Mellanox Unified Fabric Manager 1 Mellanox SX1036 40Gb Switch 36× 40Gb Ethernet Switch SX1036, 36× QSFP Interface 1 Management Network Switch Extreme Summit X440-48t-10G 2-layer Switch 48× 1Giga Switch Summit X440-48t-10G, authorized by ExtremeXOS 9 Extreme Summit X650-24X 3-layer Switch 24× 10Giga 3-layer Ethernet Switch Summit X650-24X, authorized by ExtremeXOS1 Parallel StorageDDN Parallel Storage System DDN SFA12K Storage System 1 GPU GPU Accelerator NVIDIA Tesla Kepler K20M70 MIC MIC Intel Xeon Phi 5110P Knights Corner 10 40Gb Ethernet Card MCX314A-BCBTMellanox ConnextX-3 Chip 40Gb Ethernet Card 2× 40Gb Ethernet ports, enough QSFP cables 16 SSD Intel SSD910Intel SSD910 Disk, 400GB, PCIE 80 On 8/10/2014 5:50 AM, Mark Abraham wrote: That's not what I said You can set... -npme behaves the same whether or not GPUs are in use. Using separate ranks for PME caters to trying to minimize the cost of the all-to-all communication of the 3DFFT. That's still relevant when using GPUs, but if separate PME ranks are used, any GPUs on nodes that only have PME ranks are left idle. The most effective approach depends critically on the hardware and simulation setup, and whether you pay money for your hardware. Mark On Sat, Aug 9, 2014 at 2:56 AM, Theodore Si sjyz...@gmail.com wrote: Hi, You mean no matter we use GPU acceleration or not, -npme is just a reference? Why we can't set that to a exact value? On 8/9/2014 5:14 AM, Mark Abraham wrote: You can set the number of PME-only ranks with -npme. Whether it's useful is another matter :-) The CPU-based PME offload and the GPU-based PP offload do not combine very well. Mark On Fri, Aug 8, 2014 at 7:24 AM, Theodore Si sjyz...@gmail.com wrote: Hi, Can we set the number manually with -npme when using GPU acceleration? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send
Re: [gmx-users] CUDA5.5 and ICC 14
Thanks! On 8/20/2014 8:03 PM, Mark Abraham wrote: Hi, Per Nvidia's docs and that error, CUDA 5.5 is only supported with that icc version. In practice, I understand that other icc versions work fine with GROMACS, but you would need to go and comment out that #error message. Alternatively, CUDA 6.5 supports icc 14.0 (only). Mark On Wed, Aug 20, 2014 at 8:26 AM, Theodore Si sjyz...@gmail.com wrote: Hi, I am using CUDA 5.5 and Intel ICC 14.0.1 to compile GROMACS and this happened: [ 0%] Building NVCC (Device) object src/gromacs/gmxlib/gpu_utils/ CMakeFiles/gpu_utils.dir//./gpu_utils_generated_gpu_utils.cu.o In file included from /usr/local/cuda-5.5/include/cuda_runtime.h(59), from /home/theo/gromacs-5.0/src/gromacs/gmxlib/gpu_utils/ gpu_utils.cu(0): /usr/local/cuda-5.5/include/host_config.h(72): catastrophic error: #error directive: -- unsupported ICC configuration! Only ICC 12.1 on Linux x86_64 is supported! #error -- unsupported ICC configuration! Only ICC 12.1 on Linux x86_64 is supported! ^ CMake Error at gpu_utils_generated_gpu_utils.cu.o.cmake:198 (message): Error generating /home/theo/gromacs-5.0/build/src/gromacs/gmxlib/gpu_utils/ CMakeFiles/gpu_utils.dir//./gpu_utils_generated_gpu_utils.cu.o make[2]: *** [src/gromacs/gmxlib/gpu_utils/CMakeFiles/gpu_utils.dir/./ gpu_utils_generated_gpu_utils.cu.o] Error 1 make[1]: *** [src/gromacs/gmxlib/gpu_utils/CMakeFiles/gpu_utils.dir/all] Error 2 make: *** [all] Error 2 Does this mean that I cannot make them work together? BR, Theo -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] CUDA5.5 and ICC 14
Hi, I am using CUDA 5.5 and Intel ICC 14.0.1 to compile GROMACS and this happened: [ 0%] Building NVCC (Device) object src/gromacs/gmxlib/gpu_utils/CMakeFiles/gpu_utils.dir//./gpu_utils_generated_gpu_utils.cu.o In file included from /usr/local/cuda-5.5/include/cuda_runtime.h(59), from /home/theo/gromacs-5.0/src/gromacs/gmxlib/gpu_utils/gpu_utils.cu(0): /usr/local/cuda-5.5/include/host_config.h(72): catastrophic error: #error directive: -- unsupported ICC configuration! Only ICC 12.1 on Linux x86_64 is supported! #error -- unsupported ICC configuration! Only ICC 12.1 on Linux x86_64 is supported! ^ CMake Error at gpu_utils_generated_gpu_utils.cu.o.cmake:198 (message): Error generating /home/theo/gromacs-5.0/build/src/gromacs/gmxlib/gpu_utils/CMakeFiles/gpu_utils.dir//./gpu_utils_generated_gpu_utils.cu.o make[2]: *** [src/gromacs/gmxlib/gpu_utils/CMakeFiles/gpu_utils.dir/./gpu_utils_generated_gpu_utils.cu.o] Error 1 make[1]: *** [src/gromacs/gmxlib/gpu_utils/CMakeFiles/gpu_utils.dir/all] Error 2 make: *** [all] Error 2 Does this mean that I cannot make them work together? BR, Theo -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Can we set the number of pure PME nodes when using GPUCPU?
Hi, How can we designate which CPU-only nodes to be PME-dedicated nodes? What mdrun options or what configuration should we use to make that happen? BR, Theo On 8/11/2014 9:36 PM, Mark Abraham wrote: Hi, What Carsten said, if running on nodes that have GPUs. If running on a mixed setup (some nodes with GPU, some not), then arranging your MPI environment to place PME ranks on CPU-only nodes is probably worthwhile. For example, all your PP ranks first, mapped to GPU nodes, then all your PME ranks, mapped to CPU-only nodes, and then use mdrun -ddorder pp_pme. Mark On Mon, Aug 11, 2014 at 2:45 AM, Theodore Si sjyz...@gmail.com wrote: Hi Mark, This is information of our cluster, could you give us some advice as regards to our cluster so that we can make GMX run faster on our system? Each CPU node has 2 CPUs and each GPU node has 2 CPUs and 2 Nvidia K20M Device Name Device Type Specifications Number CPU NodeIntelH2216JFFKRNodesCPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 332 Fat NodeIntelH2216WPFKRNodesCPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 256G(16×16G) ECC Registered DDR3 1600MHz Samsung Memory20 GPU NodeIntelR2208GZ4GC CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 50 MIC NodeIntelR2208GZ4GC CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 5 Computing Network SwitchMellanox Infiniband FDR Core Switch 648× FDR Core Switch MSX6536-10R, Mellanox Unified Fabric Manager 1 Mellanox SX1036 40Gb Switch 36× 40Gb Ethernet Switch SX1036, 36× QSFP Interface 1 Management Network Switch Extreme Summit X440-48t-10G 2-layer Switch 48× 1Giga Switch Summit X440-48t-10G, authorized by ExtremeXOS 9 Extreme Summit X650-24X 3-layer Switch 24× 10Giga 3-layer Ethernet Switch Summit X650-24X, authorized by ExtremeXOS1 Parallel StorageDDN Parallel Storage System DDN SFA12K Storage System 1 GPU GPU Accelerator NVIDIA Tesla Kepler K20M70 MIC MIC Intel Xeon Phi 5110P Knights Corner 10 40Gb Ethernet Card MCX314A-BCBTMellanox ConnextX-3 Chip 40Gb Ethernet Card 2× 40Gb Ethernet ports, enough QSFP cables 16 SSD Intel SSD910Intel SSD910 Disk, 400GB, PCIE 80 On 8/10/2014 5:50 AM, Mark Abraham wrote: That's not what I said You can set... -npme behaves the same whether or not GPUs are in use. Using separate ranks for PME caters to trying to minimize the cost of the all-to-all communication of the 3DFFT. That's still relevant when using GPUs, but if separate PME ranks are used, any GPUs on nodes that only have PME ranks are left idle. The most effective approach depends critically on the hardware and simulation setup, and whether you pay money for your hardware. Mark On Sat, Aug 9, 2014 at 2:56 AM, Theodore Si sjyz...@gmail.com wrote: Hi, You mean no matter we use GPU acceleration or not, -npme is just a reference? Why we can't set that to a exact value? On 8/9/2014 5:14 AM, Mark Abraham wrote: You can set the number of PME-only ranks with -npme. Whether it's useful is another matter :-) The CPU-based PME offload and the GPU-based PP offload do not combine very well. Mark On Fri, Aug 8, 2014 at 7:24 AM, Theodore Si sjyz...@gmail.com wrote: Hi, Can we set the number manually with -npme when using GPU acceleration? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo
Re: [gmx-users] Can we set the number of pure PME nodes when using GPUCPU?
Hi Mark, This is information of our cluster, could you give us some advice as regards to our cluster so that we can make GMX run faster on our system? Each CPU node has 2 CPUs and each GPU node has 2 CPUs and 2 Nvidia K20M Device Name Device Type Specifications Number CPU Node IntelH2216JFFKRNodes CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 332 Fat Node IntelH2216WPFKRNodes CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 256G(16×16G) ECC Registered DDR3 1600MHz Samsung Memory20 GPU Node IntelR2208GZ4GC CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 50 MIC Node IntelR2208GZ4GC CPU: 2×Intel Xeon E5-2670(8 Cores, 2.6GHz, 20MB Cache, 8.0GT) Mem: 64GB(8×8GB) ECC Registered DDR3 1600MHz Samsung Memory 5 Computing Network Switch Mellanox Infiniband FDR Core Switch 648× FDR Core Switch MSX6536-10R, Mellanox Unified Fabric Manager 1 Mellanox SX1036 40Gb Switch 36× 40Gb Ethernet Switch SX1036, 36× QSFP Interface 1 Management Network Switch Extreme Summit X440-48t-10G 2-layer Switch 48× 1Giga Switch Summit X440-48t-10G, authorized by ExtremeXOS 9 Extreme Summit X650-24X 3-layer Switch 24× 10Giga 3-layer Ethernet Switch Summit X650-24X, authorized by ExtremeXOS 1 Parallel StorageDDN Parallel Storage System DDN SFA12K Storage System 1 GPU GPU Accelerator NVIDIA Tesla Kepler K20M70 MIC MIC Intel Xeon Phi 5110P Knights Corner 10 40Gb Ethernet Card MCX314A-BCBT Mellanox ConnextX-3 Chip 40Gb Ethernet Card 2× 40Gb Ethernet ports, enough QSFP cables 16 SSD Intel SSD910Intel SSD910 Disk, 400GB, PCIE 80 On 8/10/2014 5:50 AM, Mark Abraham wrote: That's not what I said You can set... -npme behaves the same whether or not GPUs are in use. Using separate ranks for PME caters to trying to minimize the cost of the all-to-all communication of the 3DFFT. That's still relevant when using GPUs, but if separate PME ranks are used, any GPUs on nodes that only have PME ranks are left idle. The most effective approach depends critically on the hardware and simulation setup, and whether you pay money for your hardware. Mark On Sat, Aug 9, 2014 at 2:56 AM, Theodore Si sjyz...@gmail.com wrote: Hi, You mean no matter we use GPU acceleration or not, -npme is just a reference? Why we can't set that to a exact value? On 8/9/2014 5:14 AM, Mark Abraham wrote: You can set the number of PME-only ranks with -npme. Whether it's useful is another matter :-) The CPU-based PME offload and the GPU-based PP offload do not combine very well. Mark On Fri, Aug 8, 2014 at 7:24 AM, Theodore Si sjyz...@gmail.com wrote: Hi, Can we set the number manually with -npme when using GPU acceleration? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Can we set the number of pure PME nodes when using GPUCPU?
Hi, You mean no matter we use GPU acceleration or not, -npme is just a reference? Why we can't set that to a exact value? On 8/9/2014 5:14 AM, Mark Abraham wrote: You can set the number of PME-only ranks with -npme. Whether it's useful is another matter :-) The CPU-based PME offload and the GPU-based PP offload do not combine very well. Mark On Fri, Aug 8, 2014 at 7:24 AM, Theodore Si sjyz...@gmail.com wrote: Hi, Can we set the number manually with -npme when using GPU acceleration? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] Are there some environment variables not valid anymore?
Hi, I find this in the installation instruction part GMX 5.0： Helping CMake find the right libraries/headers/programs http://www.gromacs.org/Documentation/Installation_Instructions#TOC If libraries are installed in non-default locations their location can be specified using the following environment variables: * |CMAKE_INCLUDE_PATH|for header files * |CMAKE_LIBRARY_PATH|for libraries * |CMAKE_PREFIX_PATH|for header, libraries and binaries (e.g.|/usr/local|). However when I set the CMAKE_INCLUDE_PATH with -DCMAKE_INCLUDE_PATH I get this: 209 //No help, variable specified on the command line. 210 CMAKE_INCLUDE_PATH:UNINITIALIZED=/home/theo/myprg/vt/include/vampirtrace/ Why? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] How to profile GMX?
Hi, I would like to know what kind of profiling tools can be used with GMX? Which is the most commonly used one? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] Why is there a NxN VdW [F] on a separate line?
This is extracted from a log file of a mdrun of 512 openMP threads without GPU acceleration. Since the first line and third line both have N*N Vdw [F], does the former include the latter? As we can see, in the log file of a mdrun of 8 openMP threads without GPU acceleration, there is no standalone N*N Vdw [F], why the difference? -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] Why is there a NxN VdW [F] on a separate line?
Please compare the file 8.log https://onedrive.live.com/redir?resid=990FCE59E48164A4%212481authkey=%21AI9ThbRY_7ZgAg8ithint=file%2clog and 512.log https://onedrive.live.com/redir?resid=990FCE59E48164A4%212482authkey=%21APLkizOBzXtPHxsithint=file%2clog.Their M E G A - F L O P S A C C O U N T I N part are different as 8.log has no standalone NxN VdW [F] and NxN VdW [VF]. 512.log has the following lines. NxN VdW [F] 17.077648 563.562 0.0 NxN VdW [VF]0.002592 0.111 0.0 Why the difference? And the both have NxN Ewald Elec. + VdW [F] NxN Ewald Elec. + VdW [VF] Does NxN Ewald Elec. + VdW [F] mean NxN Ewald Elec. and NxN VdW [F]? If it is the case, why 512.log has both NxN Ewald Elec. + VdW [F] and NxN VdW [F]? On 8/5/2014 10:11 PM, Mark Abraham wrote: On Tue, Aug 5, 2014 at 4:00 AM, Theodore Si sjyz...@gmail.com wrote: This is extracted from a log file There's no data. The list cannot accept attachments, so you need to copy-paste a relevant chunk, or upload a log file to a file-sharing service. of a mdrun of 512 openMP threads without GPU acceleration. mdrun will refuse to run with 512 OpenMP threads - please report your mdrun command line rather than your mental model of it. Since the first line and third line both have N*N Vdw [F], does the former include the latter? No, but there is no line with N*N Vdw [F]. Please be precise if you are asking for detailed information. As we can see, in the log file of a mdrun of 8 openMP threads without GPU acceleration, there is no standalone N*N Vdw [F], why the difference? Can't tell, don't know what is different between the two runs. My guess is that the former run is actually running on 64 MPI ranks, each of 8 OpenMP threads, in which case you have domain decomposition per MPI rank, and in that case there are separate calls to kernels that are aimed at computing the interactions associated with atoms whose home is in different domains. You should see the ratio vary as the number of ranks varies. Mark -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] Error occurs when compiling gromacs using vtcc
Hi all, Does anyone know the instrumentation tool Vampir Trace? I am using it, and I want to instrument the gromacs code. So my cmake options are: cmake .. -DCMAKE_BUILD_OWN_FFTW=ON -DCMAKE_C_COMPILER=vtcc -DCMAKE_CXX_COMPILER=vtcxx -DGMX_MPI=on -DCMAKE_INSTALL_PREFIX=/home/theo/gmx -CUDA_TOOLKIT_ROOT_DIR=/usr/local/cuda-5.5 Vampir Trace includes some binaries, like vtcc and vtcxx above. They behave like compiler but are actually compiler wrapper. For example, you can write main.c and vtcc main.c -o main, and the binary main will be built and you can run it like a normal binary file, but during the run, some files will be generated and can be analysed latter. However, error occurs: [ 2%] Building C object src/gromacs/CMakeFiles/libgromacs.dir/__/external/tng_io/src/compression/tng_compress.c.o /home/theo/gromacs-5.0/src/external/tng_io/src/compression/tng_compress.c:25: error: expected ‘=’, ‘,’, ‘;’, ‘asm’ or ‘__attribute__’ before ‘speed’ /home/theo/gromacs-5.0/src/external/tng_io/src/compression/tng_compress.c: In function ‘quantize’: /home/theo/gromacs-5.0/src/external/tng_io/src/compression/tng_compress.c:80: warning: implicit declaration of function ‘verify_input_data’ At top level: cc1: warning: unrecognized command line option -Wno-maybe-uninitialized make[2]: *** [src/gromacs/CMakeFiles/libgromacs.dir/__/external/tng_io/src/compression/tng_compress.c.o] Error 1 make[1]: *** [src/gromacs/CMakeFiles/libgromacs.dir/all] Error 2 make: *** [all] Error 2 When I open the tng_compress.c, I find that line 25 is just a comment. 24 #define SPEED_DEFAULT 2 /* Default to relatively fast compression. For very good compression it makes sense to 25 choose speed=4 or speed=5 */ 26 Could you tell me what's going on? I'd appreciate you kind help. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [CentOS] wget: unable to resolve host address “xxxxx”
I am using CentOS 6.5, Vmware 10.0.3 It's working now. I have no idea why, since I didn't change anything. 于2014年8月1日 19:06:03,Guy Harrison写到: On Wednesday 30 July 2014 15:40:56 Theodore Si wrote: But how? Could you be more specific? I am using NAT to connect my virtual machine to the Internet. We need more information off you. What kind of vmware? What host platform? What Centos? What is your exact 'wget' command failure message? It may be a vmware problem in which case you may need to ask there. However (iirc) vmnet0 (bridged), vmnet1 (host-only) and vmnet8 (nat) are common psuedo network devices installed. Nat should work out of the box but note is effectively using whatever DNS is configured for the host. However, the fact other tools work, would, in the first instance, imply a problem with 'wget' usage. 于2014年7月30日 22:39:04,Devin Reade写到: On Jul 29, 2014, at 23:47, Gopu Krishnan gopukrishnan...@gmail.com wrote: try adding google dns 8.8.8.8 in resolv.conf His IP is in private address space; google nameservers won't help. The correct answer is to fix his local DNS config. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] How to compile APUE code on centos 6.X
It's so weird... It works now, either with #include bsd/stdlib.h or not. 于2014年7月31日 8:40:05,Jonathan Billings写到: On Thu, Jul 31, 2014 at 08:09:52AM +0800, Theodore Si wrote: I build it myself, not using rpm since it doesn't work. My OS is 32bit. It works for me with libbsd and libbsd-devel installed: $ cd /tmp $ wget http://www.apuebook.com/src.3e.tar.gz $ tar zxvf src.3e.tar.gz $ cd apue.3e $ make [multiple lines building] $ cd threads $ make gcc -ansi -I../include -Wall -DLINUX -D_GNU_SOURCE barrier.c -o barrier -L../lib -lapue -pthread -lrt -lbsd /tmp/ccmCvcBh.o: In function `thr_fn': barrier.c:(.text+0x80): undefined reference to `heapsort' collect2: ld returned 1 exit status make: *** [barrier] Error 1 $ sudo yum -y -q install libbsd libbsd-devel $ make gcc -ansi -I../include -Wall -DLINUX -D_GNU_SOURCE barrier.c -o barrier -L../lib -lapue -pthread -lrt -lbsd $ I'll admit that my OS is 64-bit, but I don't see anything in the code that would break with 32-bit. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] [CentOS 6.5]mounting hgfs shares failed
Hi all, I instaled a CentOS 6.5 in Vmware and installed the vmware-tools. Howerver, when I start up, it gives me the message that mounting hgfs shares [failed] How do I solve this, any thoughts? ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] wget: unable to resolve host address “xxxxx”
But how? Could you be more specific? I am using NAT to connect my virtual machine to the Internet. 于2014年7月30日 22:39:04,Devin Reade写到: On Jul 29, 2014, at 23:47, Gopu Krishnan gopukrishnan...@gmail.com wrote: try adding google dns 8.8.8.8 in resolv.conf His IP is in private address space; google nameservers won't help. The correct answer is to fix his local DNS config. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] wget: unable to resolve host address “xxxxx”
I tried to wget the google homepage, and It worked... So confused On Jul 31, 2014 2:02 AM, John R Pierce pie...@hogranch.com wrote: On 7/29/2014 10:11 PM, Theodore Si wrote: nameserver 192.168.80.2 is that a valid DNS server that knows how to look up the address you're trying to wget from? the wget command you're running, is it from a host on your local NAT network, or from a host on the public internet? -- john r pierce 37N 122W somewhere on the middle of the left coast ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] How to compile APUE code on centos 6.X
I build it myself, not using rpm since it doesn't work. My OS is 32bit. On Jul 31, 2014 4:21 AM, Jonathan Billings billi...@negate.org wrote: You shouldn't need the bsd/stdlib.h if the barrier.c used the function definition you described in your original message. I was able to find and build the file on CentOS 6 with the EPEL libbsd package installed. Are you sure you installed the right package? 32-bit vs. 64-bit? On July 29, 2014 11:54:01 PM EDT, Theodore Si sjyz...@gmail.com wrote: I write a .c file and #include bsd/stdlib.h and call heapsort, I get this: Apparently, heapsort can be called. 于2014年7月30日 11:46:55,Theodore Si写到: I build the libbsd from source code that I downloaded from here: http://libbsd.freedesktop.org/wiki/ I think I got all things needed, right? 于2014年7月30日 10:35:39,Theodore Si写到: I tried to install libbsd before, it didn't work.(I can execute man heapsort) Today, I installed epel-release-6-8.noarch.rpm and then yum install libbsd,(No result for man heapsort) but it still doesn't work. 于2014年7月29日 23:58:48,Jonathan Billings写到: On Tue, Jul 29, 2014 at 10:00:53PM +0800, Theodore Si wrote: Hi all, I want to compile the source code of Advanced Programming in the Unix Environment(APUE) 3rd edition, and I encountered some difficulties. After executing make, I got this message: [...] How to install libbsd to solve this problem on CentOS (this works on Ubuntu)? You need the 'libbsd' package, available in EPEL. (http://fedoraproject.org/wiki/EPEL) The 'heapsort()' function is implemented there. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos -- Jonathan Billings billi...@negate.org ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [gmx-users] Some columns in log file.
thnks a lot ! 于2014年7月29日 19:53:19,Justin Lemkul写到: On 7/28/14, 10:11 PM, Theodore Si wrote: For example, a form that explains the meanings of the all items in the log file. I found this page (http://www.gromacs.org/Documentation/Tutorials/GROMACS_USA_Workshop_and_Conference_2013/Topology_preparation,_%22What's_in_a_log_file%22,_basic_performance_improvements%3A_Mark_Abraham,_Session_1A) yesterday, it give some information about the log file, but I think it's inadequate. Computing: --- Domain decomp. Doing DD. DD comm. load DD comm. bounds DD communication stuff. Send X to PME Coordinate communication for PME. Neighbor search Self-explanatory. Comm. coord. Communicating coordinates. Force Doing forces. Wait + Comm. F Time spent waiting for, and communicating forces. PME mesh Self-explanatory. PME wait for PP Time spent on PME waiting for particle-particle interactions. Wait + Recv. PME F Time spent waiting and receiving PME forces. NB X/F buffer ops. Nonbonded forces and coordinate operations. Write traj. Self-explanatory. Update Application of all relevant updating functions. Constraints Self-explanatory. Comm. energies Communicating energy information. Rest Everything else. --- Total --- --- PME redist. X/F PME spread/gather PME 3D-FFT PME 3D-FFT Comm. PME solve --- A bunch of stuff related to how PME works, communicating data, etc. Could you explain each of these items' meaning? Especially, what does X mean? X means coordinates. F means forces. Most of these terms are technical specs that determine how much time was spent doing what during the run. Is there something in these terms that concerns you? Something indicative of bad performance? -Justin -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[CentOS] How to compile APUE code on centos 6.X
Hi all, I want to compile the source code of Advanced Programming in the Unix Environment(APUE) 3rd edition, and I encountered some difficulties. After executing make, I got this message: gcc -ansi -I../include -Wall -DLINUX -D_GNU_SOURCE barrier.c -o barrier -L../lib -lapue -pthread -lrt -lbsd /tmp/cc0gG99O.o: In function `thr_fn': barrier.c:(.text+0x6e): undefined reference to `heapsort' collect2: ld returned 1 exit status make[1]: *** [barrier] Error 1 make[1]: Leaving directory `/home/drizzlex/apue.3e/threads' make: *** [all] Error 1 In barrier.c, I found this #ifdef SOLARIS #define heapsort qsort #else extern int heapsort(void *, size_t, size_t, int (*)(const void *, const void *)); #endif If I change these code to How to install libbsd to solve this problem on CentOS (this works on Ubuntu)? ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] How to compile APUE code on centos 6.X
So all the packages of Fedora can also be used in CentOS? 于2014年7月29日 23:58:48,Jonathan Billings写到: On Tue, Jul 29, 2014 at 10:00:53PM +0800, Theodore Si wrote: Hi all, I want to compile the source code of Advanced Programming in the Unix Environment(APUE) 3rd edition, and I encountered some difficulties. After executing make, I got this message: [...] How to install libbsd to solve this problem on CentOS (this works on Ubuntu)? You need the 'libbsd' package, available in EPEL. (http://fedoraproject.org/wiki/EPEL) The 'heapsort()' function is implemented there. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] How to compile APUE code on centos 6.X
I tried to install libbsd before, it didn't work.(I can execute man heapsort) Today, I installed epel-release-6-8.noarch.rpm and then yum install libbsd,(No result for man heapsort) but it still doesn't work. 于2014年7月29日 23:58:48,Jonathan Billings写到: On Tue, Jul 29, 2014 at 10:00:53PM +0800, Theodore Si wrote: Hi all, I want to compile the source code of Advanced Programming in the Unix Environment(APUE) 3rd edition, and I encountered some difficulties. After executing make, I got this message: [...] How to install libbsd to solve this problem on CentOS (this works on Ubuntu)? You need the 'libbsd' package, available in EPEL. (http://fedoraproject.org/wiki/EPEL) The 'heapsort()' function is implemented there. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] How to compile APUE code on centos 6.X
I've installed libbsd(I think I have successed because when I execute man heapsort I see this) and I make a link /lib/libbsd.so since in the file Make.defines.linux the LDDIR is as follows: then I add this to threads/barrier.c howerver, I still get this: ? 2014/7/29 23:58, Jonathan Billings ??: On Tue, Jul 29, 2014 at 10:00:53PM +0800, Theodore Si wrote: Hi all, I want to compile the source code of Advanced Programming in the Unix Environment(APUE) 3rd edition, and I encountered some difficulties. After executing make, I got this message: [...] How to install libbsd to solve this problem on CentOS (this works on Ubuntu)? You need the 'libbsd' package, available in EPEL. (http://fedoraproject.org/wiki/EPEL) The 'heapsort()' function is implemented there. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] How to compile APUE code on centos 6.X
I build the libbsd from source code that I downloaded from here: http://libbsd.freedesktop.org/wiki/ I think I got all things needed, right? 于2014年7月30日 10:35:39,Theodore Si写到: I tried to install libbsd before, it didn't work.(I can execute man heapsort) Today, I installed epel-release-6-8.noarch.rpm and then yum install libbsd,(No result for man heapsort) but it still doesn't work. 于2014年7月29日 23:58:48,Jonathan Billings写到: On Tue, Jul 29, 2014 at 10:00:53PM +0800, Theodore Si wrote: Hi all, I want to compile the source code of Advanced Programming in the Unix Environment(APUE) 3rd edition, and I encountered some difficulties. After executing make, I got this message: [...] How to install libbsd to solve this problem on CentOS (this works on Ubuntu)? You need the 'libbsd' package, available in EPEL. (http://fedoraproject.org/wiki/EPEL) The 'heapsort()' function is implemented there. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [CentOS] How to compile APUE code on centos 6.X
I write a .c file and #include bsd/stdlib.h and call heapsort, I get this: Apparently, heapsort can be called. 于2014年7月30日 11:46:55,Theodore Si写到: I build the libbsd from source code that I downloaded from here: http://libbsd.freedesktop.org/wiki/ I think I got all things needed, right? 于2014年7月30日 10:35:39,Theodore Si写到: I tried to install libbsd before, it didn't work.(I can execute man heapsort) Today, I installed epel-release-6-8.noarch.rpm and then yum install libbsd,(No result for man heapsort) but it still doesn't work. 于2014年7月29日 23:58:48,Jonathan Billings写到: On Tue, Jul 29, 2014 at 10:00:53PM +0800, Theodore Si wrote: Hi all, I want to compile the source code of Advanced Programming in the Unix Environment(APUE) 3rd edition, and I encountered some difficulties. After executing make, I got this message: [...] How to install libbsd to solve this problem on CentOS (this works on Ubuntu)? You need the 'libbsd' package, available in EPEL. (http://fedoraproject.org/wiki/EPEL) The 'heapsort()' function is implemented there. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
[CentOS] wget: unable to resolve host address “xxxxx”
Hi all, I find that in my CentOS, which is installed in vmware, I can use yum to install software from Internet, and I can also ping websites, but I cannot download stuff using wget. I receive error msg unable to resolve host address “x”. The IP address is 192.168.80.128, and this is the content of /etc/resolv.conf # Generated by NetworkManager domain localdomain search localdomain nameserver 192.168.80.2 I'd appreciate your help. ___ CentOS mailing list CentOS@centos.org http://lists.centos.org/mailman/listinfo/centos
Re: [gmx-users] How do I get to know the meaning of the first column in log file?
Core t (s) Wall t (s)(%) Time: 2345.800 49.744 4715.7 (ns/day)(hour/ns) Performance: 69.4790.345 What does 0.345 hour/ns stand for? and the Wall time 49.77s? 于 2014/7/28 14:53, Mark Abraham 写道: I plan to put some more of this kind of documentation in the upcoming User Guide, but it isn't done yet. Number of times that section was entered, the total wall time spent in that section, and the total number of processor gigacycles spent in that section, and percentage of same. Some columns are useful for only some kinds of comparisons. Mark On Mon, Jul 28, 2014 at 5:35 AM, Theodore Si sjyz...@gmail.com wrote: Thanks a lot! But I am still confused about other things. For instance, what do count, wall t(s) G-Cycles mean? It seems that the last column is the percentage of G-Cycles. I really hope there is a place where I can find all relative information of the log file. 于 2014/7/28 11:12, Mark Abraham 写道: On Jul 28, 2014 4:53 AM, Theodore Si sjyz...@gmail.com wrote: For example, in the following form, what does Wait + Comm. F mean? Is there a webpage that explains the forms in log file? Unfortunately not (yet), but they correspond in a more-or-less clear way to the segments in manual figure 3.16. In this case, to the three boxes below Evaluate potential/forces. Significant time spent here would indicate poor balance of compute load. Mark R E A L C Y C L E A N D T I M E A C C O U N T I N G Computing: Nodes Th. Count Wall t (s) G-Cycles % - Domain decomp. 24 2 801 2.827 351.962 5.7 DD comm. load 24 2 800 0.077 9.604 0.2 DD comm. bounds 24 2 800 0.354 44.014 0.7 Neighbor search 24 2 801 1.077 134.117 2.2 Launch GPU ops. 24 2 40002 1.518 189.021 3.1 Comm. coord. 24 2 19200 2.009 250.121 4.0 Force 24 2 20001 8.478 1055.405 17.0 Wait + Comm. F 24 2 20001 1.967 244.901 4.0 PME mesh 24 2 20001 24.064 2995.784 48.4 Wait GPU nonlocal 24 2 20001 0.170 21.212 0.3 Wait GPU local 24 2 20001 0.072 8.935 0.1 NB X/F buffer ops. 24 2 78402 0.627 78.050 1.3 Write traj. 24 2 2 0.037 4.569 0.1 Update 24 2 20001 0.497 61.874 1.0 Constraints 24 2 20001 4.198 522.645 8.4 Comm. energies 24 2 801 0.293 36.500 0.6 Rest 24 1.478 184.033 3.0 - Total 24 49.744 6192.746 100.0 - - PME redist. X/F 24 2 40002 6.199 771.670 12.5 PME spread/gather 24 2 40002 7.194 895.557 14.5 PME 3D-FFT 24 2 40002 2.727 339.480 5.5 PME 3D-FFT Comm. 24 2 80004 7.460 928.742 15.0 PME solve 24 2 20001 0.434 53.968 0.9 - -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] How do I get to know the meaning of the first column in log file?
I thought that 69.479 ns/day means I can simulate 69.479 ns per day. But if as you said, I need 0.345 hour to get a simulated nanosecond, then I can only get 0.345 * 24 = 8.28 simulated nanosecond per day? Then what does 69.479 mean? And the Core t(s)? 于 2014/7/28 15:29, Mark Abraham 写道: Your run took nearly a minute, and did so at a rate that would take 0.345 hours to do a simulated nanosecond Mark On Mon, Jul 28, 2014 at 9:05 AM, Theodore Si sjyz...@gmail.com wrote: Core t (s) Wall t (s)(%) Time: 2345.800 49.744 4715.7 (ns/day)(hour/ns) Performance: 69.4790.345 What does 0.345 hour/ns stand for? and the Wall time 49.77s? 于 2014/7/28 14:53, Mark Abraham 写道: I plan to put some more of this kind of documentation in the upcoming User Guide, but it isn't done yet. Number of times that section was entered, the total wall time spent in that section, and the total number of processor gigacycles spent in that section, and percentage of same. Some columns are useful for only some kinds of comparisons. Mark On Mon, Jul 28, 2014 at 5:35 AM, Theodore Si sjyz...@gmail.com wrote: Thanks a lot! But I am still confused about other things. For instance, what do count, wall t(s) G-Cycles mean? It seems that the last column is the percentage of G-Cycles. I really hope there is a place where I can find all relative information of the log file. 于 2014/7/28 11:12, Mark Abraham 写道: On Jul 28, 2014 4:53 AM, Theodore Si sjyz...@gmail.com wrote: For example, in the following form, what does Wait + Comm. F mean? Is there a webpage that explains the forms in log file? Unfortunately not (yet), but they correspond in a more-or-less clear way to the segments in manual figure 3.16. In this case, to the three boxes below Evaluate potential/forces. Significant time spent here would indicate poor balance of compute load. Mark R E A L C Y C L E A N D T I M E A C C O U N T I N G Computing: Nodes Th. Count Wall t (s) G-Cycles % - Domain decomp. 24 2 801 2.827 351.962 5.7 DD comm. load 24 2 800 0.077 9.604 0.2 DD comm. bounds 24 2 800 0.354 44.014 0.7 Neighbor search 24 2 801 1.077 134.117 2.2 Launch GPU ops. 24 2 40002 1.518 189.021 3.1 Comm. coord. 24 2 19200 2.009 250.121 4.0 Force 24 2 20001 8.478 1055.405 17.0 Wait + Comm. F 24 2 20001 1.967 244.901 4.0 PME mesh 24 2 20001 24.064 2995.784 48.4 Wait GPU nonlocal 24 2 20001 0.170 21.212 0.3 Wait GPU local 24 2 20001 0.072 8.935 0.1 NB X/F buffer ops. 24 2 78402 0.627 78.050 1.3 Write traj. 24 2 2 0.037 4.569 0.1 Update 24 2 20001 0.497 61.874 1.0 Constraints 24 2 20001 4.198 522.645 8.4 Comm. energies 24 2 801 0.293 36.500 0.6 Rest 24 1.478 184.033 3.0 - Total 24 49.744 6192.746 100.0 - - PME redist. X/F 24 2 40002 6.199 771.670 12.5 PME spread/gather 24 2 40002 7.194 895.557 14.5 PME 3D-FFT 24 2 40002 2.727 339.480 5.5 PME 3D-FFT Comm. 24 2 80004 7.460 928.742 15.0 PME solve 24 2 20001 0.434 53.968 0.9 - -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] How do I get to know the meaning of the first column in log file?
Hi all, Could anyone explain all the meanings of these columns in log? I'd appreciate your help. Domain decomp. DD comm. load DD comm. bounds Send X to PME Neighbor search Comm. coord. Force Wait + Comm. F PME mesh PME wait for PP Wait + Recv. PME F NB X/F buffer ops. Write traj. Update Constraints Comm. energies Rest -- Total -- -- PME redist. X/F PME spread/gather PME 3D-FFT PME 3D-FFT Comm. PME solve 于2014年7月28日 15:51:15,Guillaume Chevrot写到: Hi, if you need 0.345 hour to get 1 ns, that means that you can simulate 1/0.345~2.9ns/hour, then in one day you will simulate 1/0.345*24~69.5 ns Guillaume On 07/28/2014 09:39 AM, Theodore Si wrote: I thought that 69.479 ns/day means I can simulate 69.479 ns per day. But if as you said, I need 0.345 hour to get a simulated nanosecond, then I can only get 0.345 * 24 = 8.28 simulated nanosecond per day? Then what does 69.479 mean? And the Core t(s)? 于 2014/7/28 15:29, Mark Abraham 写道: Your run took nearly a minute, and did so at a rate that would take 0.345 hours to do a simulated nanosecond Mark On Mon, Jul 28, 2014 at 9:05 AM, Theodore Si sjyz...@gmail.com wrote: Core t (s) Wall t (s)(%) Time: 2345.800 49.744 4715.7 (ns/day)(hour/ns) Performance: 69.4790.345 What does 0.345 hour/ns stand for? and the Wall time 49.77s? 于 2014/7/28 14:53, Mark Abraham 写道: I plan to put some more of this kind of documentation in the upcoming User Guide, but it isn't done yet. Number of times that section was entered, the total wall time spent in that section, and the total number of processor gigacycles spent in that section, and percentage of same. Some columns are useful for only some kinds of comparisons. Mark On Mon, Jul 28, 2014 at 5:35 AM, Theodore Si sjyz...@gmail.com wrote: Thanks a lot! But I am still confused about other things. For instance, what do count, wall t(s) G-Cycles mean? It seems that the last column is the percentage of G-Cycles. I really hope there is a place where I can find all relative information of the log file. 于 2014/7/28 11:12, Mark Abraham 写道: On Jul 28, 2014 4:53 AM, Theodore Si sjyz...@gmail.com wrote: For example, in the following form, what does Wait + Comm. F mean? Is there a webpage that explains the forms in log file? Unfortunately not (yet), but they correspond in a more-or-less clear way to the segments in manual figure 3.16. In this case, to the three boxes below Evaluate potential/forces. Significant time spent here would indicate poor balance of compute load. Mark R E A L C Y C L E A N D T I M E A C C O U N T I N G Computing: Nodes Th. Count Wall t (s) G-Cycles % - Domain decomp. 24 2 801 2.827 351.962 5.7 DD comm. load 24 2 800 0.077 9.604 0.2 DD comm. bounds 24 2 800 0.354 44.014 0.7 Neighbor search 24 2 801 1.077 134.117 2.2 Launch GPU ops. 24 2 40002 1.518 189.021 3.1 Comm. coord. 24 2 19200 2.009 250.121 4.0 Force 24 2 20001 8.478 1055.405 17.0 Wait + Comm. F 24 2 20001 1.967 244.901 4.0 PME mesh 24 2 20001 24.064 2995.784 48.4 Wait GPU nonlocal 24 2 20001 0.170 21.212 0.3 Wait GPU local 24 2 20001 0.072 8.935 0.1 NB X/F buffer ops. 24 2 78402 0.627 78.050 1.3 Write traj. 24 2 2 0.037 4.569 0.1 Update 24 2 20001 0.497 61.874 1.0 Constraints 24 2 20001 4.198 522.645 8.4 Comm. energies 24 2 801 0.293 36.500 0.6 Rest 24 1.478 184.033 3.0 - Total 24 49.744 6192.746 100.0 - - PME redist. X/F 24 2 40002 6.199 771.670 12.5 PME spread/gather 24 2 40002 7.194 895.557 14.5 PME 3D-FFT 24 2 40002 2.727 339.480 5.5 PME 3D-FFT Comm. 24 2 80004 7.460 928.742 15.0 PME solve 24 2 20001 0.434 53.968 0.9 - -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/ Support/Mailing_Lists/GMX-Users_List before posting! * Can't post
[gmx-users] How do I get to know the meaning of the first column in log file?
For example, in the following form, what does Wait + Comm. F mean? Is there a webpage that explains the forms in log file? R E A L C Y C L E A N D T I M E A C C O U N T I N G Computing: Nodes Th. Count Wall t (s) G-Cycles % - Domain decomp. 24 2 801 2.827 351.962 5.7 DD comm. load 24 2 800 0.077 9.604 0.2 DD comm. bounds 24 2 800 0.354 44.014 0.7 Neighbor search 24 2 801 1.077 134.117 2.2 Launch GPU ops. 24 2 40002 1.518 189.021 3.1 Comm. coord. 24 2 19200 2.009 250.121 4.0 Force 24 2 20001 8.478 1055.405 17.0 Wait + Comm. F 24 2 20001 1.967 244.901 4.0 PME mesh 24 2 20001 24.064 2995.784 48.4 Wait GPU nonlocal 24 2 20001 0.170 21.212 0.3 Wait GPU local 24 2 20001 0.072 8.935 0.1 NB X/F buffer ops. 24 2 78402 0.627 78.050 1.3 Write traj. 24 2 2 0.037 4.569 0.1 Update 24 2 20001 0.497 61.874 1.0 Constraints 24 2 20001 4.198 522.645 8.4 Comm. energies 24 2 801 0.293 36.500 0.6 Rest 24 1.478 184.033 3.0 - Total 24 49.744 6192.746 100.0 - - PME redist. X/F 24 2 40002 6.199 771.670 12.5 PME spread/gather 24 2 40002 7.194 895.557 14.5 PME 3D-FFT 24 2 40002 2.727 339.480 5.5 PME 3D-FFT Comm. 24 2 80004 7.460 928.742 15.0 PME solve 24 2 20001 0.434 53.968 0.9 - -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
Re: [gmx-users] How do I get to know the meaning of the first column in log file?
Thanks a lot! But I am still confused about other things. For instance, what do count, wall t(s) G-Cycles mean? It seems that the last column is the percentage of G-Cycles. I really hope there is a place where I can find all relative information of the log file. 于 2014/7/28 11:12, Mark Abraham 写道: On Jul 28, 2014 4:53 AM, Theodore Si sjyz...@gmail.com wrote: For example, in the following form, what does Wait + Comm. F mean? Is there a webpage that explains the forms in log file? Unfortunately not (yet), but they correspond in a more-or-less clear way to the segments in manual figure 3.16. In this case, to the three boxes below Evaluate potential/forces. Significant time spent here would indicate poor balance of compute load. Mark R E A L C Y C L E A N D T I M E A C C O U N T I N G Computing: Nodes Th. Count Wall t (s) G-Cycles % - Domain decomp. 24 2 801 2.827 351.962 5.7 DD comm. load 24 2 800 0.077 9.604 0.2 DD comm. bounds 24 2 800 0.354 44.014 0.7 Neighbor search 24 2 801 1.077 134.117 2.2 Launch GPU ops. 24 2 40002 1.518 189.021 3.1 Comm. coord. 24 2 19200 2.009 250.121 4.0 Force 24 2 20001 8.478 1055.405 17.0 Wait + Comm. F 24 2 20001 1.967 244.901 4.0 PME mesh 24 2 20001 24.064 2995.784 48.4 Wait GPU nonlocal 24 2 20001 0.170 21.212 0.3 Wait GPU local 24 2 20001 0.072 8.935 0.1 NB X/F buffer ops. 24 2 78402 0.627 78.050 1.3 Write traj. 24 2 2 0.037 4.569 0.1 Update 24 2 20001 0.497 61.874 1.0 Constraints 24 2 20001 4.198 522.645 8.4 Comm. energies 24 2 801 0.293 36.500 0.6 Rest 24 1.478 184.033 3.0 - Total 24 49.744 6192.746 100.0 - - PME redist. X/F 24 2 40002 6.199 771.670 12.5 PME spread/gather 24 2 40002 7.194 895.557 14.5 PME 3D-FFT 24 2 40002 2.727 339.480 5.5 PME 3D-FFT Comm. 24 2 80004 7.460 928.742 15.0 PME solve 24 2 20001 0.434 53.968 0.9 - -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
[gmx-users] Some columns in log file.
Hi all, In the log file, what do count, wall t(s) G-Cycles mean? It seems that the last column is the percentage of G-Cycles. I really hope there is a place where I can find all relative information of the log file. Thanks in advance. -- Gromacs Users mailing list * Please search the archive at http://www.gromacs.org/Support/Mailing_Lists/GMX-Users_List before posting! * Can't post? Read http://www.gromacs.org/Support/Mailing_Lists * For (un)subscribe requests visit https://maillist.sys.kth.se/mailman/listinfo/gromacs.org_gmx-users or send a mail to gmx-users-requ...@gromacs.org.
