Escali , LLC Network Router 4.4 Manual do Utilizador Página 61
- Página / 81
- Índice
- RESOLUÇÃO DE PROBLEMAS
- MARCADORES
Avaliado. / 5. Com base em avaliações de clientes
Section: 5.4 Collective operations
Scali MPI Connect Release 4.4 Users Guide 49
5.3.2 Memory consumption increase after warm-up
Remember that group operations (MPI_Comm_{create, dup, split}) may involve creating
new communication buffers. If this is a problem, decreasing chunck_size may help.
5.4 Collective operations
A collective communication is a communication operation in which a group of processes works
together to distribute or gather together a set of one or more values. Scali MPI Connect uses
a number of different approaches to implement collective operations. Through environment
variables the user can control which algorithm the application uses.
Consider the Integer Sort (IS) benchmark in NPB (NAS Parallel Benchmarks). When running on
ten processes on 5 nodes over Gigabit Ethernet (mpimon -net smp,tcp bin/is.A.16.scampi
-- r1 2 r2 2 r3 2 r4 2 r5 2) the resulting performance is:
Mop/s total = 34.05
Mop/s/process = 2.13
Extracting the MPI profile of the run can be done as follows:
user% export SCAMPI_TRACE="-f arg;timing"
user% mpimon bin/is.A.16.scampi -- $ALL2 > trace.out
And running the output through scanalyze yields the following:
MPI Call <128 128-1k 1-8k 8-32k 32-256k 256k-1M >1M
MPI_Send 0.00 0.00 0.00 0.00 0.00 0.00 0.00
MPI_Irecv 0.00 0.00 0.00 0.00 0.00 0.00 0.00
MPI_Wait 0.69 0.00 0.00 0.00 0.00 0.00 0.00
MPI_Alltoall 0.14 0.00 0.00 0.00 0.00 0.00 0.00
MPI_Alltoallv 11.20 0.00 0.00 0.00 0.00 0.00 0.00
MPI_Reduce 1.04 0.00 0.00 0.00 0.00 0.00 0.00
MPI_Allreduce 0.00 0.00 15.63 0.00 0.00 0.00 0.00
MPI_Comm_size 0.00 0.00 0.00 0.00 0.00 0.00 0.00
MPI_Comm_rank 0.00 0.00 0.00 0.00 0.00 0.00 0.00
MPI_Keyval_free 0.00 0.00 0.00 0.00 0.00 0.00 0.00
The MPI_Alltoallv uses a high fraction of the total execution time. The communication time is
the sum of all used algorithms and the total timing may depend on more than one type of
communication. If one type or a few operations dominate the time consumption, they are
promising candidates for tuning/optimization.
Note: Please note that the run time selectable algorithms and their values may vary on
different Scali MPI Connect release versions. For information on which algorithms that are
selectable at run time and their valid values, set the environment variable
SCAMPI_ALGORITHM and run an example application:
# SCAMPI_ALGORITHM=1 mpimon /opt/scali/examples/bin/hello -- localhost
This will produce a listing of the different implementations of particular collective MPI calls.
For each collective operation a listing consisting of a number and a short description of the
algoritmn is produced, e.g., for MPI_Alltoallv() the following:
SCAMPI_ALLTOALLV_ALGORITHM alternatives
0 pair0
1 pair1
2 pair2
3 pair3
- Scali MPI Connect 1
- Users Guide 1
- Software release 4.4 1
- Acknowledgement 2
- Table of contents 13
- Chapter 1 Introduction 17
- 1.2 Support 18
- 1.3 How to read this guide 19
- 1.6 Typographic conventions 21
- 1.5 Terms and conventions 21
- 2.2 SMC network devices 24
- 2.2.1 Network devices 25
- 2.2.2 Shared Memory Device 25
- 2.2.3 Ethernet Devices 25
- 2.2.3.2 DET 26
- 2.2.3.3 Using detctl 26
- 2.2.3.4 Using detstat 26
- 2.2.4 Myrinet 27
- 2.2.5 Infiniband 27
- 2.2.6 SCI 28
- 2.3.1 Channel buffer 28
- 2.3.2 Inlining protocol 29
- 2.3.3 Eagerbuffering protocol 29
- 2.3.4 Transporter protocol 29
- 2.5 MPI-2 Features 30
- Section: 2.5 MPI-2 Features 32
- 3.2 Compiling and linking 33
- 3.2.2 Compiler support 34
- 3.2.3 Linker flags 34
- 3.3.1 Naming conventions 35
- 3.3.2.1 Basic usage 36
- 3.3.2.4 Standard input 37
- 3.3.2.5 Standard output 37
- 3.3.2.7 Network options 38
- 3.3.3 mpirun - wrapper script 39
- 3.7 Debugging and profiling 41
- 3.9.3 Using MPI_Bsend() 44
- 3.9.5 Unsafe MPI programs 45
- 3.9.6 Name space pollution 45
- 3.10.2 Fatal errors 45
- 3.11 Mpimon options 46
- Section: 3.11 Mpimon options 48
- 4.1 Example 49
- 4.2 Tracing 50
- 4.3 Timing 53
- 4.4 Using the scanalyze 55
- 5.3 Benchmarking 60
- 5.4 Collective operations 61
- Appendix A Example MPI code 63
- Appendix B Troubleshooting 66
- B-1.4 General problems 67
- C-8 License options 70
- C-9 Scali kernel drivers 71
- C-10 Uninstalling SMC 71
- Section: 73
- D-1 Bracket expansion 74
- D-2 Grouping 74
- List of figures 78
Milwaukee sds 5363-21 manuals
Owner’s manuals and user’s guides for Power tools Milwaukee sds 5363-21.
We providing 1 pdf manuals Milwaukee sds 5363-21 for download free by document types: Operator's Manual
Comentários a estes Manuais