CN103714006A - Performance test method of Gromacs software - Google Patents
Performance test method of Gromacs software Download PDFInfo
- Publication number
- CN103714006A CN103714006A CN201410006794.2A CN201410006794A CN103714006A CN 103714006 A CN103714006 A CN 103714006A CN 201410006794 A CN201410006794 A CN 201410006794A CN 103714006 A CN103714006 A CN 103714006A
- Authority
- CN
- China
- Prior art keywords
- compiling
- gromacs software
- compiler
- gromacs
- software
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Debugging And Monitoring (AREA)
- Stored Programmes (AREA)
Abstract
The invention provides a performance test method of Gromacs software. Different compilers, a math library and compiler parameters are used for compiling the Gromacs software. Simulation test is carried out on the Gromacs software after compiling, compiling performance data are collected, compiling performance comparison is executed, the compiler with the best compiling performance, the math library and compiling parameters are selected, nPME parameters are adjusted again, and subsequent optimization is carried out on the Gromacs software. The performance test method can improve operation efficiency of the Gromacs software.
Description
Technical field
The present invention relates to field of computer technology, be specifically related to a kind of performance test methods of Gromacs software.
Background technology
GROMACS software is a general routine package of increasing income that is mainly used to carry out molecular dynamics simulation, can simulate the Newton's equation group of hundreds of to millions of scale particIe system motions.It is mainly used to simulated albumin matter and the such biomolecule of lipid, also can be used for research such as abiotic molecular systems such as polymkeric substance.GROMACS software support is common nearly all molecular dynamics algorithm at present, comprises various heating baths, presses bath and electrostatic interaction to calculate etc.GROMACS software application is very extensive, but the continuous increase along with computer size, computing platform is constantly updated and simulated system constantly increases, everybody attention further to counting yield, but at present also nobody the optimization of gromacs is proposed to the comprehensive scheme of system.
Summary of the invention
The present invention proposes a kind of performance test methods of Gromacs software, and to improve software operation efficiency, described method comprises:
Step 1: configure hardware platform, compiler, math library and compiling parameter;
Step 2: use different compilers, math library and the compiling parameter of configuration to compile Gromacs software, the Gromacs software after compiling is carried out simulation test and collected compilation performance data;
Step 3: carry out compilation performance contrast, select compiler, math library and the compiling parameter of compilation performance optimum to arrange;
Step 4: adjust nPME parameter, carry out the follow-up optimization of Gromacs software.
The invention has the beneficial effects as follows and improve Gromacs software operation efficiency.
Accompanying drawing explanation
Fig. 1 is the process flow diagram of the performance test methods of a kind of Gromacs software of proposing of the present invention.
Fig. 2 is that the method that proposes of the present invention is to nPME parameter optimization schematic diagram.
Embodiment
With reference to Fig. 1, Fig. 1 shows the process flow diagram of the method for the present invention's proposition, and described method comprises the steps:
Step 1: configure hardware platform, compiler, math library and compiling parameter.
Step 2: use different compilers and compiling parameter configuration to compile Gromacs software.In this step, can use different compilers and compiling parameter to compile Gromacs software, for example, first use GNU compiler, fft math library compiles Gromacs software, and compiling parameter Uses Defaults; Compiling completes, and carries out molecular dynamics simulation; Collect compilation performance data; And then use INTEL compiler, and MKL math library is carried out the compiling of Gromacs software, and compiling parameter is as follows:
-DCMAKE_C_COMPILER=mpiicc
-DCMAKE_CXX_COMPILER=mpiicpc
-DGMX_MPI=ON
-DGMX_GPU=OFF
-DGMX_DOUBLE=ON
-DBUILD_SHARED_LIBS=OFF
-DGMX_PREFER_STATIC_LIBS=ON
-DCMAKE_C_FLAGS="-ip?xAVX-funroll-all-loops-static-O3"
-DCMAKE_EXE_LINKER_FLAGS="-L/opt/intel/composer_xe_2013.1.117/mkl/lib/intel64/-lmkl_intel_lp64-lmkl_intel_thread-lmkl_core;
After having compiled, carry out molecular dynamics simulation; Collect compilation performance data.
Step 3: carry out compilation performance contrast, select compiler, math library and the compiling parameter of compilation performance optimum to arrange.For example, by the compilation performance difference of contrast GNU/INTEL compiler, the GROMACS performance that GNU compiles out, compared with low 10% left and right of intel compiler, is used INTEL compiler, MKL math library and corresponding compiling parameter to carry out the test of follow-up operation parameter optimization.
Step 4: adjust nPME parameter, carry out the follow-up optimization of Gromacs software.In GROMACS, conventionally use PME(Particle-mesh Ewald), P3M(Particle-Particle-Particle Mesh), the methods such as Worf are processed the overall situation and are interacted, while adopting Ewald to add the long range interaction of processing with method, gromacs software can be specified and is specifically designed to the number that carries out PME computing node by nPME parameter, all the other nodes carry out PP(particle-particle) process, the computing node that participates in calculating will be divided into two parts like this, so arranging of nPME parameter affects on operational efficiency larger.Gromacs software is meeting default allocation nPME when operation, and we float in its left and right with reference to default value when testing, and through testing in a large number the selective rules of summing up nPME, Fig. 2 shows this process.
Certainly; the present invention also can have other various embodiments; in the situation that not deviating from spirit of the present invention and essence thereof; those of ordinary skill in the art are when making according to the present invention various corresponding changes and distortion, but these corresponding changes and distortion all should belong to the protection domain of claim of the present invention.
Claims (3)
1. a performance test methods for Gromacs software, is characterized in that, described method comprises:
Step 1: configure hardware platform, compiler, math library and compiling parameter;
Step 2: use different compilers, math library and the compiling parameter of configuration to compile Gromacs software, the Gromacs software after compiling is carried out simulation test and collected compilation performance data;
Step 3: carry out compilation performance contrast, select compiler, math library and the compiling parameter of compilation performance optimum to arrange;
Step 4: adjust nPME parameter, carry out the follow-up optimization of Gromacs software.
2. the method for claim 1, is characterized in that:
Described compiler comprises GUN and INTEL compiler, and described math library comprises FFT and MKL.
3. the method for claim 1, is characterized in that:
Described step 4 is specially: select different nPME parameters, sum up the selective rules of nPME through a large amount of tests, complete the follow-up optimization of Gromacs software.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410006794.2A CN103714006B (en) | 2014-01-07 | 2014-01-07 | Performance test method of Gromacs software |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201410006794.2A CN103714006B (en) | 2014-01-07 | 2014-01-07 | Performance test method of Gromacs software |
Publications (2)
Publication Number | Publication Date |
---|---|
CN103714006A true CN103714006A (en) | 2014-04-09 |
CN103714006B CN103714006B (en) | 2017-05-24 |
Family
ID=50407003
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201410006794.2A Active CN103714006B (en) | 2014-01-07 | 2014-01-07 | Performance test method of Gromacs software |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN103714006B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108664729A (en) * | 2018-05-10 | 2018-10-16 | 深圳晶泰科技有限公司 | A kind of GROMACS cloud computings flow control method |
WO2019134322A1 (en) * | 2018-05-10 | 2019-07-11 | 深圳晶泰科技有限公司 | Gromacs cloud computing process control method |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050273310A1 (en) * | 2004-06-03 | 2005-12-08 | Newburn Chris J | Enhancements to performance monitoring architecture for critical path-based analysis |
CN1983209A (en) * | 2005-12-14 | 2007-06-20 | 中兴通讯股份有限公司 | System and method for automatically testing software unit |
CN103309715A (en) * | 2013-06-26 | 2013-09-18 | 曙光信息产业(北京)有限公司 | Method for installing molecular dynamics application program in high-performance trunking system |
-
2014
- 2014-01-07 CN CN201410006794.2A patent/CN103714006B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20050273310A1 (en) * | 2004-06-03 | 2005-12-08 | Newburn Chris J | Enhancements to performance monitoring architecture for critical path-based analysis |
CN1983209A (en) * | 2005-12-14 | 2007-06-20 | 中兴通讯股份有限公司 | System and method for automatically testing software unit |
CN103309715A (en) * | 2013-06-26 | 2013-09-18 | 曙光信息产业(北京)有限公司 | Method for installing molecular dynamics application program in high-performance trunking system |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108664729A (en) * | 2018-05-10 | 2018-10-16 | 深圳晶泰科技有限公司 | A kind of GROMACS cloud computings flow control method |
WO2019134322A1 (en) * | 2018-05-10 | 2019-07-11 | 深圳晶泰科技有限公司 | Gromacs cloud computing process control method |
US10922453B2 (en) | 2018-05-10 | 2021-02-16 | Shenzhen Jingtai Technology Co., Ltd. | GROMACS cloud computing process control method |
CN108664729B (en) * | 2018-05-10 | 2021-11-23 | 深圳晶泰科技有限公司 | GROMACS cloud computing flow control method |
Also Published As
Publication number | Publication date |
---|---|
CN103714006B (en) | 2017-05-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Atzeni et al. | ARCHER: effectively spotting data races in large OpenMP applications | |
Juve et al. | Characterizing and profiling scientific workflows | |
Asadchev et al. | New multithreaded hybrid CPU/GPU approach to Hartree–Fock | |
Wei et al. | DLVM: A modern compiler infrastructure for deep learning systems | |
CN104536898B (en) | The detection method of c program parallel regions | |
Mantovani et al. | Performance and energy consumption of HPC workloads on a cluster based on Arm ThunderX2 CPU | |
AR093834A1 (en) | SYSTEMS AND METHODS TO TEST A FINANCIAL MARKETING SYSTEM | |
Ma et al. | LAMP: data provenance for graph based machine learning algorithms through derivative computation | |
Krommydas et al. | On the characterization of opencl dwarfs on fixed and reconfigurable platforms | |
KR20120100421A (en) | System and method for fluid simulation using interaction between particles | |
US9396095B2 (en) | Software verification | |
US8959498B2 (en) | Parallelization method, system and program | |
Gonzalez-de-Aledo et al. | An optimization approach for agent-based computational models of biological development | |
CN103714006A (en) | Performance test method of Gromacs software | |
CN102722438A (en) | Kernel debugging method and equipment | |
CN102521135B (en) | The method of testing of linear system and device | |
CN103778001A (en) | Analysis method for degree of parallelism of simulation task on basis of DAG (Directed Acyclic Graph) | |
US9038045B2 (en) | Unified parallel C work-sharing loop construct transformation | |
Blazewicz et al. | A massive data parallel computational framework for petascale/exascale hybrid computer systems | |
US8555030B2 (en) | Creating multiple versions for interior pointers and alignment of an array | |
US8661424B2 (en) | Auto-generation of concurrent code for multi-core applications | |
Wei et al. | DLVM: A modern compiler framework for neural network DSLs | |
Campora Perez et al. | An efficient low‐rank Kalman filter for modern SIMD architectures | |
US20120102507A1 (en) | Streaming programming generator | |
Afzal et al. | Making applications faster by asynchronous execution: Slowing down processes or relaxing MPI collectives |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |