--------------020304090509080005080001-- From bioclusters@bioinformatics.org Thu Jul 1 14:49:23 2004 From: bioclusters@bioinformatics.org (Joe Landman) Date: Thu, 01 Jul 2004 09:49:23 -0400 Subject: [Bioclusters] updated NCBI 2.2.9 RPMs In-Reply-To: <40E4122E.8080606@accessgate.net> References: <1088533045.16839.553.camel@protein.scalableinformatics.com> <40E4122E.8080606@accessgate.net> Message-ID: <1088689762.2787.30.camel@protein.scalableinformatics.com> On Thu, 2004-07-01 at 09:31, Doug Shubert wrote: > Hello Joe, > Ran your static 2.9.9.-5 build against a dynamic build from src. > results are bellow. > > Please elaborate on the static compile flags you used. Hi Doug: You should be able to pull the compilation options from the patches on the site. Depending upon the target, We tried to use the best performing options that resulted in correct execution. We have some unit tests in place, and this is why when users have indicated some problems we are quite interested to get their tests to see if we can uncover the problems (or if they are problems with our build). If you are doing an Opteron run, I believe we are using -O3 -msse2 -m64 as our compilation options. On our machines, the static and dynamic RPMs give within a few percent of each other in terms of performance on BBS. By default, NCBI uses -mpentiumpro on x86 class machines, including Opteron. Our patches fix this, and have worked well since 2.2.6. What is interesting is that our 64 bit binary is about 20-30% faster than a similarly compiled 32 bit binary (built on same hardware, with same compilers, but using the -m32 option and no -msse2). The moral of this anecdote (which has some significant amount of analysis behind it, as will be seen in a few weeks) is that the 64 bit binaries seem to be better on average, and in number of cases, significantly better. We are getting on average about 28% improvement in various BLAST cases, about 30% in HMMer, and about 30-40% in a number of chemistry applications. You can run the 32 bit versions, but the 64 bit seem to be (with few exceptions) better performing. Note: all the tests were done using GCC. We are going to try other compilers as well, and report on them. All RPMs, including the source are available from http://downloads.scalableinformatics.com/downloads/ncbi/ . The patches should be there as well, though you can always pull down the source RPM, install it, and run rpmbuild -bp --target=x86_64 ncbi-toolkit.spec (or ncbi-toolkit-static.spec if you wish to use that). Joe > Doug > > SI-Static build 2.2.9-5 > 1,1,1564 > 2,2,1543 > 2,1,1572 > 3,4,3095 > 3,2,3097 > 3,3,3114 > 3,1,3117 > 4,1,796 > 5,1,1554 > 5,2,1558 > 6,1,789 > > 2.2.9 src compiled dynamic > 1,1,1660 > 2,1,1661 > 2,2,1665 > 3,2,3300 > 3,4,3311 > 3,1,3329 > 3,3,3329 > 4,1,855 > 5,1,1680 > 5,2,1682 > 6,1,851 > > > Joe Landman wrote: > > Folks: > > > > Fixed a number of problems with the spec file and with the build. > > Also set up a working static build, so you won't need the same libraries > > as I have. Makes the binaries quite a bit larger though. We are > > working on getting the source RPM for this and the dynamic build out > > later on today. You can pull the latest build from > > http://downloads.scalableinformatics.com/downloads/ncbi/ (as usual). > > These use the tarball as released from NCBI on 15-June. > > > > Note that this build corrects the performance problem on Opteron that > > some people had reported to us. > > > > Joe > > > > -- Joseph Landman, Ph.D Scalable Informatics LLC, email: landman@scalableinformatics.com web : http://scalableinformatics.com phone: +1 734 612 4615 From bioclusters@bioinformatics.org Fri Jul 2 22:38:30 2004 From: bioclusters@bioinformatics.org (Malcolm Sole) Date: Fri, 02 Jul 2004 17:38:30 -0400 Subject: [Bioclusters] Oscar installation Message-ID: <40E5D5D6.5010304@broadleafservices.com> Hi Chris I am also trying to get systemimager to work with a 3Ware controller (and shortly with a SATA drive) I have been through the manual but cant find what you are referring to (I am somewhat new at this kernel stuff). Any pointer you could give me would be most well received. Even to pointing me to the right piece of documentation. TIA Malcolm In the above message you mention The systemimager 3.x autoinstall kernel that comes down via the PXE netboot did not see the drives so it bombed out with lots of errors. Teaching systemimager to download and 'insmod' the 3ware driver was pretty darn easy. From bioclusters@bioinformatics.org Tue Jul 6 16:35:15 2004 From: bioclusters@bioinformatics.org (Ognen Duzlevski) Date: Tue, 6 Jul 2004 15:35:15 +0000 (UTC) Subject: [Bioclusters] opteron vs xeon quad cpu boxes In-Reply-To: <20040628162432.GE73468@iib.unsam.edu.ar> References: <1088435619.3470.58.camel@protein.scalableinformatics.com> <20040628162432.GE73468@iib.unsam.edu.ar> Message-ID:Folks: Fixed a number of problems with the spec file and with the build. Also set up a working static build, so you won't need the same libraries as I have. Makes the binaries quite a bit larger though. We are working on getting the source RPM for this and the dynamic build out later on today. You can pull the latest build from http://downloads.scalableinformatics.com/downloads/ncbi/ (as usual). These use the tarball as released from NCBI on 15-June. Note that this build corrects the performance problem on Opteron that some people had reported to us. Joe
*******************************************************************= *************
**** &nb= sp; &= nbsp; &n= bsp; &nb= sp; &nbs= p; ****
** = ; = CALL FOR PARTICIPATION = **
**** &nb= sp; &nbs= p; = ; = &= nbsp; ****
*******************************************************************= *************
Cluster 2004
The 2004 IEEE International Conference on Cluster = Computing
September 20-23, 2004
Town and Country Hotel, San Diego, California, USA
http://grail.sdsc.edu/cluster2004/<= /p>
*******************************************************************= *************
**** &nb= sp; &nbs= p; = ; = &= nbsp; ****
** EARLY REGISTRATION = DEADLINE: August 23, = 2004 &nbs= p; **
** Register online at http://grail.sdsc.edu/cluster2004/registration.html **
**** &= nbsp; &n= bsp; &nb= sp; &nbs= p; = ; ****
*******************************************************************= *************
The Cluster 2004 conference, to be held in = beautiful San Diego, provides an open forum for researchers, practitioners, and = users to present and discuss issues, directions, and results that will shape the = future of cluster computing. The Cluster series of conferences is one of the = flagship events sponsored by the IEEE Task Force on Cluster Computing (TFCC) = since its inception in 1999. The competition among refereed papers was = particularly strong this year, with 48 papers being selected as full papers from the = 150 submitted papers. Besides the technical paper presentation, there will = be three exciting keynote speakers, four tutorials, one workshop and exhibits to = be arranged during the conference period.
San Diego is California's second largest city = and the United States' seventh largest, San Diego boasts a citywide = population of nearly 1.3 million residents and more than 2.8 million residents = countywide. Within its borders of 4,200 sq. miles, San Diego County encompasses 18 incorporated cities and numerous other charming neighborhoods and = communities, including downtown's historic Gaslamp Quarter, Little Italy, Coronado, = La Jolla, Del Mar, Carlsbad, Escondido, La Mesa, Hillcrest, Barrio Logan, = Chula Vista and more. Known for its near-idyllic climate, 70 miles of pristine beaches and dazzling array of world-class family attractions, including = the World-Famous San Diego = Zoo = and Wild Animal = Park, SeaWorld San = Diego = and LEGOLAND = California, San = Diego offers a wide variety of things to see and do, appealing to guests from around = the world.
Cluster 2004 will be held at = the Town and Country = Resort, located = in Mission Valley, just 10 minutes from the airport, right next to the San Diego = Trolley, the Fashion Valley Mall, and the Riverwalk Golf course. Reservations = should be made with the hotel directly. The conference rate of US$129.00 (which = you can obtain by mentioning the "IEEE Cluster 2004" conference) will = be guaranteed until August 28, 5PM PST. After this date rooms reserved for Cluster'04 will be given to anyone on a first-come first-serve basis. = The Hotel online registration is available from http://grail.sdsc.edu/cluster2004/ where = you can also find the online conference registration and the Cluster 2004 advanced = program.
We look forward meeting you in San = Diego!
*******************************************************************= *************
**** &nb= sp; &nbs= p; = ; = &= nbsp; ****
** = ; = PROGRAM HIGHLIGHTS &= nbsp; &n= bsp; **
**** &nb= sp; &nbs= p; = ; = &= nbsp; ****
*******************************************************************= *************
KEYNOTES
--------
* Using = Clusters in Biomedical Image Analysis -- S. Pieper (Harvard University, USA)
* The = decision to migrate Riken's Vector Supercomputing Facility to = Clusters -- R. Himeno(Riken, = Japan)
* The = Cluster Agenda: first Achieve World Domination, then Kick = Ass, T. Sterling (Caltech, = USA)
TUTORIALS
---------
* = Parallel I/O: Lessons learnt in the last 20 years -- Toni Cortes, = Universitat Polit=E8cnica de Catalunya (UPC, CEPBA-IBM Research Institute = (CIRI)
* = Building Highly Available HPC Clusters with HA-OSCAR -- Chokchai Leangsuksun, Louisana Tech University, and Ibrahim Haddad, Ericsson = Research
* State = of InfiniBand in Designing HPC Clusters, Storage/File Systems, and = Datacenters -- D.K. Panda, Ohio State University
* MPI = Tuning with Intel=A9 Trace Analyzer and Intel=A9 Trace Collector -- Ray = Asbury and Michael Wrinn, Intel
WORKSHOP
--------
=B7 Workshop = on Remote Direct Memory Access (RDMA): Application, Implementation, and = Technologies (RAIT)
TECHNCIAL SESSIONS
-------------------
* Optimizing with MPI
* Scheduling
* Parallel I/O and Efficient = Communications
* Fault Tolerance
* Collective Communication = Optimizations
* Novel Storage = Architectures
* Networking
* Distributed Shared Memory
* SSI and Location Aware = Algorithms
* Systems Analysis
* Grid Systems
* Performance Analysis
* Visualization and = Simulation
* Systems Management
* Application Techniques =
-------------------------------------------------------------------= ---------