PlumpOS is an openMosix-node-on-a-cdrom. It was originally based off ClumpOS and continues to try to be an easy/small/powerful way of implementing openMosix nodes.
Cluster I/O fencing agent for virtual machines designed to support various methods of guest-to-host communication and management frameworks.
XLB is a high performance HTTP load balancer. connection management, caching, ssl, scripting. 300 mbit/sec / 4000 reqs/sec takes 30% cpu on a 2GhZ Xeon. connection pooling to backend servers reduces memory and cpu usage on backends.
dsh is a program which runs a single command on multiple computers at the same time. It was designed as a cluster tool for beowulf-style supercomputers.
Glusterfs filesystem for Rocks Cluster
Glusterfs is a high performance filesystem from Red Hat. This Roll integrates Glusterfs in Rocks
lustre client roll for rocks clusters
This roll provides rpm packages to mount the lustre filesystem on the frontends and compute nodes of a rocks cluster. Please contact me if you need this roll for another kernel and/or lustre version.
XCPU is a suite of tools for cluster management. It includes utilities for job starting, cluster resource management, scalable distribution of boot images as well as creation and control of virtual machines in a cluster environment.
High performance system software for clusters.
A distributed, peer-to-peer, server-less communication & messaging solution for JAVA (JMS), C++ and .NET applications, integrates with JBoss, WebLogic & WebSphere. Offers guaranteed delivery, security and transactions. Supports TCP, SSL & HTTP protocols.
Fura is a self-contained grid middleware that allows the grid enablement and distribution of applications on heterogeneous computational resources. Fura features a web-based GUI, wizard-guided installation and configuration, and Web Services compliance.
Thanks to carefully thought out architecture an emulator for all hardware (i386 i860 sparc hppa SGI Mips Alpha etc) on any hardware. A lot of new theories have gone into this trying to fix old problems.
Secure, effective and universal Virtual Operating System implementation; completely new VM design; simple but powerful rIPC mechanisms - see http://argante.sourceforge.net/advocacy.txt for more information.
Netbootmgr provides a GUI and CLI for managing the next boot action for network booted nodes of a cluster. The GUI includes node status information and provides an interface to cluster node power control.
Alternative implementation of Spring framework remoting subsystem that supports different schemes of remote communication between client and server (1-1, 1-many, dynamic services discovering).
- DSI is a security framework addressing security needs of carrier-grade Linux clusters. - DigSig is a linux kernel module checking RSA signatures of ELF executables at run-time. **Those projects are no longer maintained**
Batch system monitoring and archiving
Job Monarch is an addon to the Ganglia Monitoring System that provides batch job monitoring and archiving plus a graphical overview of clusters and assorted batch systems. Fully supported batch system: Torque, PBS and SLURM. Experimental: LSF, SGE
GlobalEnvironment provides globally shared key-value pairs for different Java threads, JVMs, J2EE (EJB and JSP), Java and non-java applications. See other <a href="http://acelet.com/super">J2EE tools</a> other information.
JCGrid is an easy-to-use tools for real-world grid-computing. It requires only to install Java Runtime and you can easy split your cpu-intensive task among several Workstations/PC/Mac/etc. It includes support for POVRay and MandelGrid as usage examples.
The Linux InfiniBand Project is a collection of sub-projects focused around providing InfiniBand drivers for existing 2.4 based kernels. See Home Page for details.
LinuxTuples is an open-source implementation of David Gelernter's idea of a tuple space, designed to run on a networked cluster of Linux machines.
Fully packaged linux distribution to provide internet access and resource management to small and medium companies.
Momoko is a framework for developing distributed, multi-user, network-aware applications. It includes facilities for storing objects in a tree structure, a shell for manipulating the object database, orthogonal persistence, scripting support, and a trans
NChord is a C# implementation of the Chord distributed hash table. The project provides a library containing the routing, lookup, and maintenance routines specified in the MIT Chord paper, and is quite stable including under heavy load and churn.
OpenMonet is a distributed, flexible monitoring tool developed in Erlang, suitable for network monitoring, host performance tracking, and for the instrumentation of complex systems.
Dataflow parallel programming language for clusters
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.