Artifact Submission: HPC MaBoSS

This is a replication package containing code and experimental results related to the paper titled: MaBoSS for HPC environments: Implementations of the continuous time Boolean model simulator for large CPU clusters and GPU accelerators.

Overview

The artifact comprises the following directories:

• benchmark -- benchmarking scripts
• data -- the (bnd, cfg) model pairs on which the benchmarks were run
• plots -- scripts for generating results plots
• presented-results -- plots (including some that were not included in the paper), CSV data files with measurements and R script that generated the plots from the data
• repos -- all the implementations of cross-correlation
  • MaBoSS.GPU -- GPU code (located under src)
  • MaBoSS -- CPU and MPI code (located under engine/src)

Detailed artifact contents

repo/MaBoSS.GPU/src directory contains the source files to the CUDA kernels. Notably, the runtime compilation optimization can be viewed in kernel_compiler.h/cpp and generator.h/cpp source files, the simulation part in jit_kernels/simulation.cu and the stats computation part in jit_kernels/final_states.cu and jit_kernels/window_average_small.cu.

MPI implementation of CPU MaBoSS code can be found by searching for the MPI_COMPAT keyword in repo/MaBoSS/engine/src directory.

Requirements for running the experiments

Hardware requirements:

• A CUDA-compatible GPU
• A (homogeneous) CPU cluster

Software requirements:

• CUDA toolkit 12.2 or later and appropriate driver
• an MPI implementation, such as mpich
• cmake 3.18 or later
• flex and bison libraries
• R software for plotting the graphs (see details below)

Let us present a few commands for your convenience that will allow you to set up the environment quickly:

Installing all dependencies (except CUDA and GPU driver) on Debian/Ubuntu:

sudo apt-get update && apt-get install -y g++ cmake r-base flex bison mpich

Installing all dependencies (except CUDA and GPU driver) on RHEL-like distribution:

sudo dnf install -y cmake gcc-c++ R flex bison mpich

R packages necessary for generating the plots:

R -e "install.packages(c('ggplot2', 'cowplot', 'sitools', 'viridis', 'dplyr'), repos='https://cloud.r-project.org')"

Running the experiments

Our experiments are designed to provide a comprehensive analysis of the aforementioned algorithms running various combinations of parameters computing different sizes of input instances. Therefore, the overall duration of running the experiments is quite long (around 2 to 3 days on our GPU cluster and MareNostrum SC).

To provide a swift way to check the reproducibility of our experiments, we prepared a special script that runs only a subset of the benchmarks.

Kick the tires:

Just to see whether the code is working, run the following from the root directory:

./kick-the-tires.sh

The script should take about 10 minutes to finish. The script runs a subset of GPU experiments.

After the script runs, it will generate results in a CSV format in the results directory. It should contain 2 CSV files for CPU and GPU benchmark respectively. Each CSV file contains self-documenting headers. Finally, the plotting script is executed generating a single plot in the plots directory. More details on how the CSV results rows are processed into plots can be found in the plots/plots-fast.R script.

The generated plot file will be named real.pdf and it shows the comparison of CPU and GPU runtime on a small cellcycle model.

Complete set of measurements:

To run the complete benchmark, execute

./run-all.sh

Measured results

The measured data and plots were stored in the presented-results directory. The directory also contains plots.R script, which was used to plot the data. It can be executed by Rscript plots.R if you wish to re-generate the plots from the data yourself. Here we describe each figure in the presented-results/plots directory:

Plot	Description	Figure number in paper
real.pdf	GPU benchmark on real-world dataset	Figure 1
sizek_mpi.pdf	MPI benchmark on real-world dataset	Figure 4
synth_mpi_speedup.pdf	MPI strong-scaling speedup on synthetic dataset	Figure 5
synth_mpi.pdf	MPI wall time on synthetic dataset	not included
nodes.pdf	GPU benchmark on synthetic dataset	Figure 2
nodes-compilation-big-NVIDIA RTX 3070 Laptop GPU.pdf	Laptop GPU runtime compilation benchmark	not included
nodes-compilation-big-NVIDIA Tesla A100 GPU.pdf	Datacenter-grade GPU runtime compilation benchmark	Figure 3