Download presentation
Presentation is loading. Please wait.
1
Parallel Computing @ ISDS Chris Hans 29 November 2004
2
Parallel Computing @ ISDS2 Organization Basics of Parallel Computing Structural Computational Coding for Distributed Computing Examples Resources at Duke CSEM Cluster
3
29 November 2004Parallel Computing @ ISDS3 Basics of Distributed Computing Online Tutorial (Livermore Nat. Lab.) http://www.llnl.gov/computing/tutorials/parallel_comp/ Serial Computing: one computer, one CPU Parallel Computing: multiple computers working at the same time
4
29 November 2004Parallel Computing @ ISDS4 Various Setups Collection of Workstations PVM (Parallel Virtual Machine) R (rpvm, Rmpi, SNOW) LAM-MPI (Local Area Multicomputer) Matlab Dedicated Cluster Master/Slave model MPI
5
29 November 2004Parallel Computing @ ISDS5 Network Layout Basic layout: Each node has: CPU(s), memory
6
29 November 2004Parallel Computing @ ISDS6 Designing a Parallel Program Do the nodes need to interact? “Embarrassingly Parallel” Very little communication -- Monte Carlo “Shamelessly Parallel” Communication Needed -- Heat diffusion models -- Spatial models
7
29 November 2004Parallel Computing @ ISDS7 Message Passing Interface “MPI” Standard Easy to use functions that manage the communication between nodes.
8
29 November 2004Parallel Computing @ ISDS8 Master/Slave Model Organize the layout: Slaves
9
29 November 2004Parallel Computing @ ISDS9 Master/Slave Model “Master” divides the task into pieces… …while slaves “listen” to the network, waiting for work. Master sends out the work to be done… …slaves do the work… …while the Master waits for the answers. Slaves return the results.
10
29 November 2004Parallel Computing @ ISDS10 Example: Monte Carlo
11
29 November 2004Parallel Computing @ ISDS11 Example: Monte Carlo Master Slave 1 Draw N/3 values Compute Return Slave 2 Same Slave 3 Same
12
29 November 2004Parallel Computing @ ISDS12 Code Write ONE program Same for master and slaves Run program on EACH node Each program has to figure out if it’s the master or a slave MPI
13
29 November 2004Parallel Computing @ ISDS13 Pseudo Code LOAD DATA; GET ID; IF(ID==MASTER) { MASTER(); } ELSE { SLAVE(); }... LOAD DATA; GET ID; IF(ID==MASTER) { MASTER(); } ELSE { SLAVE(); }... LOAD DATA; GET ID; IF(ID==MASTER) { MASTER(); } ELSE { SLAVE(); }... LOAD DATA; GET ID; IF(ID==MASTER) { MASTER(); } ELSE { SLAVE(); }... MasterSlave 1Slave 2Slave 3
14
29 November 2004Parallel Computing @ ISDS14 MASTER { // Find # of nodes GET NP; for(i in 1:NP) { “Tell process i to compute the mean for 1000 samples” } RET=RES=0; // Wait for results WHILE(RET<NP){ ANS = RECEIVE(); RES+=ANS/NP; RET++; } SLAVE { ANS = 0; // Wait for orders // Receive NREPS for(i in 1:NREPS) { ANS += DRAW(); } SEND(ANS/NREPS); RETURN TO MAIN; }; SLAVE { ANS = 0; // Wait for orders // Receive NREPS for(i in 1:NREPS) { ANS += DRAW(); } SEND(ANS/NREPS); RETURN TO MAIN; }; SLAVE { ANS = 0; // Wait for orders // Receive NREPS for(i in 1:NREPS) { ANS += DRAW(); } SEND(ANS/NREPS); RETURN TO MAIN }; MasterSlave 1Slave 2Slave 3
15
29 November 2004Parallel Computing @ ISDS15 PRINT RES; RETURN TO MAIN ; }... FINALIZE(); }... FINALIZE(); }... FINALIZE(); }... FINALIZE(); } MasterSlave 1Slave 2Slave 3 Pseudo Code
16
29 November 2004Parallel Computing @ ISDS16 Example: C++ Code Large dataset N=40, p = 1,000 One outcome variable, y Calculate R 2 for all 1-var regression parallel_R2.cpp
17
29 November 2004Parallel Computing @ ISDS17 CSEM Cluster @ Duke Computational Science, Engineering and Medicine Cluster http://www.csem.duke.edu/Cluster/clustermain.htm Shared machines Core Nodes Contributed Nodes
18
29 November 2004Parallel Computing @ ISDS18 CSEM Cluster Details 4 Dual processing head nodes 64 Dual processing shared nodes Intel Xeon 2.8 GHz 40 Dual processing “stat” nodes Intel Xeon 3.1 GHz 161 Dual processing “other” nodes Owners get priority
19
29 November 2004Parallel Computing @ ISDS19
20
29 November 2004Parallel Computing @ ISDS20 Using the Cluster Access limited ssh –l cmh27 cluster1.csem.duke.edu Data stored locally on cluster Compile using mpicc, mpif77, mpif90 Cluster uses SGE Queuing System
21
29 November 2004Parallel Computing @ ISDS21 Queuing System Submit your job with requests memory usage number of CPUs (nodes) Assigns nodes and schedules job Least-loaded machines fitting requirements Jobs run outside of SGE are killed
22
29 November 2004Parallel Computing @ ISDS22 Compiling Linked libraries, etc… g++ -c parallel_R2.cpp -I/opt/mpich-1.2.5/include -L/opt/mpich-1.2.5/lib mpicc parallel_R2.o -o parallel_R2.exe -lstdc++ -lg2c -lm
23
29 November 2004Parallel Computing @ ISDS23 Submitting a Job Create a queue script: #!/bin/tcsh # #$ -S /bin/tcsh -cwd #$ -M username@stat.duke.edu -m b #$ -M username@stat.duke.edu -m e #$ -o parallel_R2.out -j y #$ -pe low-all 5 cd /home/stat/username/ mpirun -np $NSLOTS -machinefile $TMPDIR/machines parallel_R2.exe Email me when the job is submitted and when it finishes Output file Priority and number of nodes requested Your Home Directory on CSEM Cluster
24
29 November 2004Parallel Computing @ ISDS24 Submitting a Job Type: [cmh27@head1 cmh27]$ qsub parallel_R2.q Your job 93734 ("parallel_R2.q") has been submitted. Monitoring http://clustermon.csem.duke.edu/ganglia/
25
29 November 2004Parallel Computing @ ISDS25 Downloadable Files You can download the slides, example C++ code and queuing script at: http://www.isds.duke.edu/~hans/tutorials.html
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.