Open Source Benchmark Software

Collectl is a light-weight performance monitoring tool capable of reporting interactively as well as logging to disk. It reports statistics on cpu, disk, infiniband, lustre, memory, network, nfs, process, quadrics, slabs and more in easy to read format.

Distributed session caching tools and APIs, primarily for SSL/TLS servers though perhaps useful for other (non-SSL/TLS) circumstances. Also includes a self-contained network abstraction library (libnal), and the sslswamp SSL/TLS benchmark/test utility.

PARUS is a data-flow parallel programming language that allows to build parallel programs for clusters and MPP multiprocessors. The data-flow graph is automatically converted to the C++/MPI source and linked with the libparus runtime library. Also there are available tools for benchmarking cluster interconnect and visualize it.

This is an opensource project for the calculation of the mathematical constant Pi. If you are a mathematician or programer please contribute ideas or code for this project. Non professionals are also welcome to contribute.

Cbench is intended as a relatively straightforward toolbox of tests, benchmarks, applications, utilities, and framework to hold them together with the goal to facilitate scalable testing, benchmarking, and analysis of a Linux parallel compute cluster.

Cbench is intended to be a relatively straightforward collection of tests, benchmarks, applications, utilities, and framework to hold them together with the goal to facilitate scalable testing and benchmarking of a Linux parallel compute cluster.

This is a software tool that automates the process of benchmarking high performance computer clusters by auto analyzing the hardware of the cluster and configuring and compiling specialized benchmarks (HPC Challenge, IOzone, Netperf)

High Performance Computing test

This project is to develop a versatile benchmark for High Performance Computing (HPC) network environments, currently the Linux-based Beowulf systems. The goal is to examine the TCP/UPD/MPI communications and their impact to system kernel.

A test suite to MPI-2 and threads.

MQbench is a graphical message passing (mpi, mpich) benchmark, which involves Qt for graphical view of different MPI Calls. MQbench has the abiltiy to benchmark the communication behaviour between different groups of clusternodes with a single click.

A collection of performance analysis software that can assist with a variety of techniques useful for application software optimization and benchmarking. The current release provides support for analysis with hardware performance counters on Linux.

A slightly modified base package Slackware image suitable for VM performance testing. Project Goals: Development of production testing protocols Base Image Base +Apache Base + MySQL Base + node.js Base + AoE Standard Full Slackware Current VM image The goal of the Slackless project is really two-fold. To provide high quality virtual images of the original Slackware operating system suitable for use and testing in virtual environments in an effort to introduce users new to Slackware to the advantages of the K.I.S.S. development philosophy. It's longer range goal is to produce a procedure for producing minimal functional server images for use in production environments. The goal is to minimize the image size and resource consumption while maintaining the stability and functionality that makes Slackware famous in an effort to provide implementers of virtual environments and option that allows them to install more server images on existing hardware platforms.

SysUnit is a JUnit framework for distributed testing, system testing and integration testing. SysUnit is ideal for testing highly distributed software such as clusters of web applications, web services or MOM based software

sparse linear algebra library, with emphasis on preconditioned Krylov subspace methods for the solution of linear systems of equations

coNCePTuaL is a toolset for rapidly generating portable, readable, and reproducible network-performance tests. coNCePTuaL can perform the equivalent of many pages of C code with just a few mouse clicks or lines of code in a domain-specific language.