Performance results of LeanMD on the Joliot-Curie supercomputer using different load balancing algorithms

This dataset contains the raw output files generated from the execution of LeanMD on the Joliot-Curie supercomputer (20 SKL Irene nodes), the scripts used to generate them, and the scripts used to parse these results for statistical analysis and plotting. <strong>Software information</strong>: OS: Red Hat Enterprise Linux 7.6 OpenMPI: version 2.0.4 Compilers: C/C++ Intel 17.0.6.256 Charm++ version: v6.9.0-rc3, build mpi-linux-x86_64 --with-production LeanMD source: https://charm.cs.illinois.edu/gerrit/gitweb?p=benchmarks/leanmd.git Additional load balancers source: https://github.com/viniciusmctf/packing-schemes/tree/packs_2019-v1 Charm++, LeanMD, and the load balancers were compiled with -O3 <strong>File information</strong>: The raw result files are organized in four directories (oct18, oct23, oct24, and oct24_2).<br> Each directory contains the results of one batched execution in the supercomputer.<br> Each batch is composed of 10 repetitions of a set of experiments.<br> Each set of experiments includes different load balancing algorithms and different problem sizes.<br> Each set is randomly ordered to avoid interference coming from a specific order of execution.<br> Each raw file contains the appended output of the application and its load balancer for all 10 repetitions.<br> The name of the files indicate the load balancer and size of the problem.<br> For instance, `PackStealLB.240` means that the application was run with PackStealLB and the problem size parameter is 240. <strong>Problem sizes</strong>: 80: 80×11×5 cells of dimensions 15×15×30 120: 120×11×5 cells of dimensions 15×15×30 160: 160×11×5 cells of dimensions 15×15×30 240: 240×11×5 cells of dimensions 15×15×30 320: 320×11×5 cells of dimensions 15×15×30 Each execution of LeanMD ran for 301 iterations with load balancing calls at iterations 40, 140, and 240. <strong>Raw output files</strong>: Each raw output file starts with a Charm++ header providing information on the execution. For each step of the application, its execution time in ms is provided. Load balancing calls usually provide information about their start time, end time, and duration. Depending on the load balancer, more information is provided. Each raw file contains ten executions of the application with a given load balancer and input size. <strong>Generating plots</strong>: The analysis of the results can be done by running the Jupyter notebook named "Analysis of load balancing results.ipynb"本数据集包含在约里奥-居里超级计算机（Joliot-Curie supercomputer，20个SKL Irene节点）上运行LeanMD所生成的原始输出文件、用于生成这些文件的脚本，以及用于解析结果以开展统计分析与绘图的脚本。<strong>软件信息</strong>：操作系统：Red Hat Enterprise Linux 7.6；OpenMPI版本：2.0.4；编译器：C/C++ Intel 17.0.6.256；Charm++版本：v6.9.0-rc3，编译配置为mpi-linux-x86_64 --with-production；LeanMD源码地址：https://charm.cs.illinois.edu/gerrit/gitweb?p=benchmarks/leanmd.git；额外负载均衡器源码地址：https://github.com/viniciusmctf/packing-schemes/tree/packs_2019-v1；Charm++、LeanMD及负载均衡器均以-O3编译参数进行编译。<strong>文件信息</strong>：原始结果文件分为四个目录（oct18、oct23、oct24及oct24_2）。<br>每个目录对应一次超级计算机上的批量执行任务。<br>每一批次包含一组实验的10次重复运行。<br>该组实验涵盖不同的负载均衡算法与不同的问题规模。<br>每组实验的执行顺序均经过随机化处理，以避免特定执行顺序带来的干扰。<br>每个原始文件包含该应用程序及其负载均衡器针对全部10次重复运行的追加输出。<br>文件名可指示所使用的负载均衡器与问题规模。<br>例如，`PackStealLB.240`表示该应用程序使用PackStealLB负载均衡器运行，且问题规模参数为240。<strong>问题规模</strong>：80：80×11×5个计算单元，单元尺寸为15×15×30；120：120×11×5个计算单元，单元尺寸为15×15×30；160：160×11×5个计算单元，单元尺寸为15×15×30；240：240×11×5个计算单元，单元尺寸为15×15×30；320：320×11×5个计算单元，单元尺寸为15×15×30。每次LeanMD运行均执行301次迭代，并在第40、140及240次迭代时调用负载均衡功能。<strong>原始输出文件</strong>：每个原始输出文件开头包含Charm++头信息，用于记录执行相关内容。针对应用程序的每个步骤，均会提供其执行时间（单位：毫秒）。负载均衡调用通常会记录其开始时间、结束时间与持续时长。根据负载均衡器的不同，还会提供更多相关信息。每个原始文件包含给定负载均衡器与输入规模下的10次应用程序运行结果。<strong>绘图生成</strong>：可通过运行名为"Analysis of load balancing results.ipynb"的Jupyter笔记本对结果进行分析。