1 / 26

Matrix Transpose Results with Hybrid OpenMP / MPI

Matrix Transpose Results with Hybrid OpenMP / MPI. O. Haan Gesellschaft für wissenschaftliche Datenverarbeitung Göttingen, Germany ( GWDG ). SCICOMP 2000, SDSC, La Jolla. Overview. Hybrid Programming Model Distributed Matrix Transpose Performance Measurements Summary of Results.

osgood
Télécharger la présentation

Matrix Transpose Results with Hybrid OpenMP / MPI

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Matrix Transpose Resultswith Hybrid OpenMP / MPI O. Haan Gesellschaft für wissenschaftliche DatenverarbeitungGöttingen, Germany( GWDG ) SCICOMP 2000, SDSC, La Jolla

  2. Overview • Hybrid Programming Model • Distributed Matrix Transpose • Performance Measurements • Summary of Results O. Haan, Matrix Transpose Results, SCICOMP 2000

  3. Architecture of Scalable Parallel Computers Two level hierarchy • cluster of SMP nodes distributed memory high speed interconnect • SMP nodes with multiple processors shared memory bus or switch connected O. Haan, Matrix Transpose Results, SCICOMP 2000

  4. Programming Models • message passing over all processors MPI implementation for shared memory multiple access to switch adaptersSP: 4-way Winterhawk2+8-way Nighthawk- • shared memory over all processors virtual global address space SP: - • hybrid message passing - shared memory message passing between nodes shared memory within nodesSP:+ O. Haan, Matrix Transpose Results, SCICOMP 2000

  5. Hybrid Programming Model SPMD programwith MPI tasksOpenMP threadswithin each taskcommunicationbetween MPI tasks O. Haan, Matrix Transpose Results, SCICOMP 2000

  6. Example of Hybrid Program program hybrid_example include “mpif.h“ com = MPI_COMM_WORLD call MPI_INIT(ierr) call MPI_COMM_SIZE(com,nk,ierr) call MPI_COMM_RANK(com,my_task,ierr) kp = OMP_GET_NUM_PROCS() !$OMP PARALLEL PRIVATE(my_thread) my_thread = OMP_GET_THREAD_NUM() call work(my_thread,kp,my_task,nk,thread_res) !$OMP END PARALLEL do i = 0 , kp-1 node_res = node_res + thread_res(i) end do call MPI_REDUCE(node_res,glob_res,1, : MPI_REAL,MPI_SUM,0,com,ierr) call MPI_FINALIZE(ierr) stop end O. Haan, Matrix Transpose Results, SCICOMP 2000

  7. Hybrid Programming vs.Pure Message Passing + • works on all SP configuration • coarser internode communication granularity • faster intranode communication - • larger programming effort • additional synchronization steps • reduced reuse of cached data the net score depends on the problem O. Haan, Matrix Transpose Results, SCICOMP 2000

  8. Distributed Matrix Transpose O. Haan, Matrix Transpose Results, SCICOMP 2000

  9. 3-step Transpose n1 x n2 matrix A( i1 , i2 ) --> n2 x n1 matrix B( i2 , i1 ) decompose n1, n2 in local and global parts: n1 = n1l * np n2 = n2l * np write matrices A, B as 4-dim arrays: A( i1l , i1g , i2l ; i2g ) , B( i2l , i2g , i1l ; i1g ) step 1 : local reorder A( i1l , i1g , i2l ; i2g ) -> a1( i1l , i2l , i1g ; i2g ) step 2 : global reorder a1( i1l , i2l , i1g ; i2g ) -> a2( i1l , i2l , i2g ; i1g ) step 3 : local transpose a2( i1l , i2l , i2g ; i1g ) -> B( i2l , i2g , i1l ; i1g ) O. Haan, Matrix Transpose Results, SCICOMP 2000

  10. Local Steps: Copy with Reorder • data in memory:speed limited by performance of bus and memory subsystemsWinterhawk2 : all processors share the same bus bandwidth : 1.6 GB/s • data in cache:speed limited by processor performanceWinterhawk2 : one load plus one store per cyclebandwidth : 8 MB / (1/375) s = 3 GB / s O. Haan, Matrix Transpose Results, SCICOMP 2000

  11. Copy: Data in Memory O. Haan, Matrix Transpose Results, SCICOMP 2000

  12. Copy : Prefetch O. Haan, Matrix Transpose Results, SCICOMP 2000

  13. Copy : Data in Cache O. Haan, Matrix Transpose Results, SCICOMP 2000

  14. Global Reorder a1( *, *, i1g ; i2g ) -> a2( * , * , i2g ; i1g ) global reorder on np processors in np steps p0 p1 p2 step0 step1 step2 O. Haan, Matrix Transpose Results, SCICOMP 2000

  15. Performance Modelling Hardware model: nk nodes with kp procs each np = nk * kp is total procs count Switch model: nk concurrent links between nodes latency tlat , bandwidth c execution model for Hybrid: reorder on nk nodes: nk steps with n1*n2 / nk**2 data per node execution model for MPI: reorder on np processors: np steps with n1*n2 / np**2 data per node switch links shared between kp procs O. Haan, Matrix Transpose Results, SCICOMP 2000

  16. Performance Modelling Hybrid timing model: MPI timing model: O. Haan, Matrix Transpose Results, SCICOMP 2000

  17. Timing of Global Reorder (internode part) O. Haan, Matrix Transpose Results, SCICOMP 2000

  18. Timing of Global Reorder (internode part) O. Haan, Matrix Transpose Results, SCICOMP 2000

  19. Timing of Global Reorder O. Haan, Matrix Transpose Results, SCICOMP 2000

  20. Timing of Transpose O. Haan, Matrix Transpose Results, SCICOMP 2000

  21. Scaling of Transpose O. Haan, Matrix Transpose Results, SCICOMP 2000

  22. Timing of Transpose Steps O. Haan, Matrix Transpose Results, SCICOMP 2000

  23. Summary of Results: Hardware • Memory access in Winterhawk2 is not adaquate:copy rate of 400 MB/s = 50 Mwords/s peak CPU rate of 6000 Mflops/sa factor of 100 between computational speed and memory speed • Sharing of switch link by 4 processors degrades communication speed:bandwidth smaller by more than a factor of 4 ( factor of 4 expected )latency larger by nearly a factor of 4 ( factor of 1 expected ) O. Haan, Matrix Transpose Results, SCICOMP 2000

  24. Summary of Results: Hybrid vs. MPI • hybrid OpenMP / MPI programming is profitable for distributed matrix tranpose :1000 x 1000 matrix on 16 nodes : 2.3 times faster10000 x 10000 matrix on 16 nodes : 1.1 times faster • Competing influences :MPI programming enhances use of cached dataHybrid programming has lower communication latency and coarser communication granularity O. Haan, Matrix Transpose Results, SCICOMP 2000

  25. Summary of Results: Use of Transpose in FFT 2-dim complex array of size Execution time on nk nodes : where r : computational speed per nodec : transpose speed per node effective execution speed per node : O. Haan, Matrix Transpose Results, SCICOMP 2000

  26. Summary of Results: Use of Transpose in FFT- Example SP r = 4 * 200 Mflop/s = 800 Mflop/sc depends on n, nk and programming model nk = 16 n = 10**6 10**9 hybrid c = 5.6 7.8 Mword/sMPI c = 2.5 7.0 Mword/s effective execution speed per node hybrid = 208 338 Mflop/s MPI = 108 317 Mflop/s O. Haan, Matrix Transpose Results, SCICOMP 2000

More Related