2What is MCSR? Who is MCSR? What Does MCSR Do? Who Does MCSR Serve? What Kinds of Accounts?Why Does Mississippi Need Supercomputers?What Kinds of Research?What Kinds of Instruction?What Kinds of Workshops?How Much Does it Cost?What Kinds of Software?What Supercomputers and Clusters?
3What is MCSR? Mississippi Center for Supercomputer Research Established in 1987 by the Mississippi LegislatureMission: Enhance Computational Research Climate at Mississippi’s 8 Public Universitiesalso: Support High Performance Computing (HPC) Education in Mississippi
4What Does MCSR Do? We make Mississippi scientists: - more competitive for federal grants- more productive in researchWe provide extraordinary learning opportunities for Mississippi college students- instructional accounts- computing workshops- helpdesk support
5Who Does MCSR Serve?MCSR serves faculty, researchers, and students at all of Mississippi’s 8 public universitiesAlcorn State UniversityJackson State UniversityMississippi Sate UniversityMississippi Valley State UniversityThe University of Southern MississippiDelta State UniversityThe University of MississippiMississippi University for Women
7What Types of Computing Access is Available? Research Accounts- provided for faculty and student “researchers”- good for the duration of employment or enrollmentInstructional Accounts- provided at an instructor’s request- for all students enrolled in a semester course- valid for the duration of the semester- can be “converted” to research account
12Why Do Mississippi Researchers Need Supercomputers? Economics: researchers in a poor state like Mississippi can still make a big splash.Computational simulations are faster, cheaper, and safer than laboratory experiments alone.
13What Kinds of Research @ MCSR? 90% of MCSR calculations are computational chemistryCleanup of high explosive materials.Design of high energy density rocket fuelsThe chemical underpinnings of high powered lasersMutation studies of enzyme activityDesigning weather-proofing coatings for machineryOther AreasHurricane forecastingBlast resistant CoatingsBetter 3-D imaging for diagnosing brain tumors
14What Types of Workshops by MCSR? MCSR consultants taught over 140 free seminars in FY08.Over 60 training topics available, and growing.Fixed schedule or on-demand.Unix/programming, Math Software, Stats Software, Computational Chemistry Software
16Do Researchers and Students Pay to Use MCSR? No. MCSR services are provided at no cost to the individual, department, or institution.Funded researchers may ask for priority access.Mississippi researchers may claim the value of MCSR computing services received as an in-kind contribution from their institution when seeking federal grants.
27Introduction to Parallel Programming at MCSR Message Passing ComputingProcesses communicate via calls to message passing library routinesProgrammers “parallelize” algorithm and add message callsAt MCSR, this is via MPI programming with C or FortranSweetgum – Origin 2800 Supercomputer (128 CPUs)Mimosa – Beowulf Cluster with 253 NodesRedwood – Altix 3700 Supercomputer (224 CPUs)Sequoia – Altix XE 310 InfiniBand Cluster (176 cores)Shared Memory ComputingThreads coordinate/communicate results via shared memory variablesCare must be taken not to modify the wrong memory areasAt MCSR, this is via OpenMP programming with C or Fortran on sweetgum, redwood, or sequoia
29What MCSR Systems for USM Class Accounts SweetgumMPI or OpenMP1 to 16 CPUsUp to 900mb per CPUPBS scripts preferred#PBS –l ncpus=4Interactive computations will be killed after 30 minutesQueues: SM-4P, SM-8P, MM-8P, MM-16PProcessors: Mix of 195 MHz and 300 MHzO/S: Irix (like Unix)Compilers: SGI’s Fortran, C/C++, GNU C/C++, w/ SGI MPT
32What MCSR Systems for USM Class Accounts MimosaMPI1 to 18 nodes400 GB Memory per nodePBS Scripts Only (no interactive jobs allowed)#PBS –l nodes=4Queues: MCSR-CAProcessors: single 1.4 GHz P4/nodeO/S: SUSE Linux 10.3Compilers: Portland Group (PGI) Fortran, C/C++ w/mpichqstat –f (to find out what nodes your job is running on)
35What MCSR Systems for USM Class Accounts SequoiaOpenMP (multiple processors on the same node)MPI (multiple processors on the same or different nodes)Hybrid (OpenMP within node, MPI across nodes)1 to 4 nodes, 1 to 8 CPUs per nodePBS Queues: SM-4P (for up to 4 CPUs on 1 node)MCSR-Test (up to 8 CPUs on each of 4 nodes)PBS Scripts Only (no interactive jobs allowed)#PBS –l nodes=4:ppn=8 (to run on all 8 CPUs of all 4 nodes)#PBS –l ncpus=8:select=16 GB Memory per node (2 GB per CPU)qstat –f (to find out what nodes your job is running on)
38Sequoia for USM Class Accounts To run on 4 nodes, and 8 processors per node (32 processes)#PBS –l nodes=4:ppn=8To run on 2 nodes, and 4 processors per node (8 processes)#PBS –l nodes=2:ppn=4To run on 1 node, and up to 8 processors (OpenMP)#PBS –l nodes=1:ppn=8To run on 8 processors, regardless of number of nodes#PBS –l ncpus=8To run 8 processors, with preferences about node placement#PBS –l place=scatter (distribute across as many nodes as can)#PBS –l place=pack (pack processes onto as few nodes as can)#PBS –l place=free (place processes on first available processors)
39Parallel EfficiencyNSF Expiration Sep 30, 2004; HPVCI Expiration Date Dec 31, 2002.