I have installed hpcc package to benchmark my system. Its description is as follows:
Description-en: HPC Challenge benchmark
The High Performance Computing (HPC) Challenge benchmark runs a suite of 7 tests that measure the performance of CPU, memory and network for HPC clusters. Amongst others, it includes the High-Performance LINPACK (HPL) benchmark, used by the Top500 ranking ().
It has executable named hpcc and placed in /usr/bin/hpcc.
If I run it - I get error message:
$ hpcc
HPL WARNING from process # 0, on line 313 of function HPL_pdinfo:
>>> cannot open file hpccinf.txt <<<How to correctly run hpcc and where can I get hpccinf.txt file?
1 Answer
According to man hpcc
The High Performance Computing (HPC) Challenge benchmark runs a suite of tests that measure the performance of CPU, memory and network for HPC clusters. hpcc takes its parameters from a hpccinf.txt file. An example can be found in
/usr/share/doc/hpcc/examples/_hpccinf.txt.
So we need to copy /usr/share/doc/hpcc/examples/_hpccinf.txt to current directory with name hpccinf.txt, edit it and run it with mpirun.openmpi hpcc as described in /usr/share/doc/hpcc/README.Debian:
HPC Challenge Benchmark for Debian
Please read
/usr/share/doc/hpcc/README.txt.gz, especially section 'Runtime configuration'.An
hpccinf.txtinput file is provided as/usr/share/doc/hpcc/examples/_hpccinf.txt. Copy it into your current dir, tune it and launchhpccusingmpirun.openmpi: $mpirun.openmpihpcc-- Lucas Nussbaum Sat, 13 Jun 2009 16:04:17 +0200
So we have two options:
use default
hpccinf.txtfrom repository and run benchmarkcp /usr/share/doc/hpcc/examples/_hpccinf.txt hpccinf.txt mpirun.openmpi -np $(nproc) hpccThe results will be saved in
hpccoutf.txtfile.customize
hpccinf.txtfor modern systems with 4-8 cores (solving matrix with 10000x10000 dimmensions):cat << EOF > hpccinf.txt HPLinpack benchmark input file Innovative Computing Laboratory, University of Tennessee HPL.out output file name (if any) 6 device out (6=stdout,7=stderr,file) 1 # of problems sizes (N) 10000 Ns 1 # of NBs 128 NBs 0 PMAP process mapping (0=Row-,1=Column-major) 1 # of process grids (P x Q) 1 Ps 1 Qs 16.0 threshold 1 # of panel fact 2 PFACTs (0=left, 1=Crout, 2=Right) 1 # of recursive stopping criterium 4 NBMINs (>= 1) 1 # of panels in recursion 2 NDIVs 1 # of recursive panel fact. 1 RFACTs (0=left, 1=Crout, 2=Right) 1 # of broadcast 1 BCASTs (0=1rg,1=1rM,2=2rg,3=2rM,4=Lng,5=LnM) 1 # of lookahead depth 1 DEPTHs (>=0) 0 SWAP (0=bin-exch,1=long,2=mix) 1 swapping threshold 1 L1 in (0=transposed,1=no-transposed) form 1 U in (0=transposed,1=no-transposed) form 0 Equilibration (0=no,1=yes) 8 memory alignment in double (> 0) EOFThen run benchmark and interpret the results
mpirun.openmpi -np $(nproc) hpcc && grep Gflops$ -A3 hpccoutf.txtExamples for 64-bit Ubuntu 16.04.4 LTS:
+------------------------+---------|-----------+----|----| | CPU | Threads | Gflops | Ps | Qs | +------------------------+---------+-----------|----+----+ | Intel i7-740QM | 8 | 16.4 | 1 | 1 | | Intel i7-920 | 8 | 28.1 | 2 | 2 | | Intel i7-4790 | 8 | 137.1 | 1 | 1 | | Intel i7-3537U | 4 | 14.3 | 2 | 2 | | AMD A4-4000 | 2 | 6.6 | 2 | 1 | | Intel Core 2 Duo E8300 | 2 | 16.2 | 2 | 1 | | Intel Pentium G3420 | 2 | 26.1 | 2 | 1 | | Raspberry Pi 3B+ | 4 | 1.9 | 1 | 1 | +------------------------+---------+-----------|----+----+
Note: if have Intel you can use also their optimized LINPACK benchmark. Its results is +25% higher.
1
