openMosix is a a set of extensions to the standard Linux kernel allowing you to build a cluster of out of off-the-shelf PC hardware. openMosix scales perfectly up to thousands of nodes. You do not need to modify your applications to benefit from your cluster (unlike PVM, MPI, Linda, etc.). Processes in openMosix migrate transparently between nodes and the cluster will always auto-balance.
openMosixLOAF (openMosix Linux-On-A-Floppy) is a single floppy openMosix Linux mini-distro designed for harddiskless systems or HPC clusters where only CPU and memory is needed. It has drivers for all PCI Ethernet NICs that the Linux kernel has support for (10/100/1000) and comes on only one bootable 1.6MB 3½" floppy diskette, ready to be a node in your HPC cluster.
openSSI webView is a simple and easy-to-use openSSI cluster monitoring system. Its goal is to provide a quick overview of the cluster state by graphing vital functions and graphically representing key figures. It allows the cluster administrator to keep an eye on the cluster health and usage rate, to quickly view each node state and load, and to watch, and even migrate, users' processes all across the cluster.
p-run is a utility that runs a program, script, or series of commands on a large number of hosts in parallel. It simplifies management of machines in computer labs. It uses ssh, and supports password and key-based authentication. p-run uses libexpect, and will run in parallel across as many processes as specified. When run on a large number of hosts, its parallel nature significantly speeds up administration.
pconsole is a parallel console tool, and is meant as an interactive administrative shell tool for clusters. It allows you to connect to each node of your cluster simultaneously, and you can type your administrative commands in a specialized window that 'multiplies' the input to each of the connections you have opened.
penbw (pen backend watcher) is a small addon for pen, a high-availability load balancer for TCP. penbw monitors and blacklists a (http-)backend automatically in case of a failure. pen itself is able to detect if a backend is down and will blacklist it on the fly. But if, for example, the backend is still accepting connections and simply sending back bogus or nothing, pen will not detect that, as pen doesn't care about application-level protocols. By contrast, penbw will be effective in such a situation.