Distributed operating systems and its elements:

1. OpenVMS Galaxy http://www.openvms.compaq.com/availability/galaxy.html. Commercial distributed system on the base of OpenVMS with appropriate hardware support. It enables full support of distributed resources of several computers.
   
2. UnixWare Non Stop Cluster http://www.sco.com/products/clustering/nscwhtpaper/ . Commercial distributed system, its main goal is to provide reliability and performance increasing.
   
3. openMOSIX (Multicomputer Operating System for unIX) http://www.openmosix.org/ http://www.openmosix.org/ System of process migration with network usage for Linux. It allows load distribution on all processes and optimal usage of all nodes RAM. Platform Intel x86 is supported. It is enough stable for practical usage. Open source code. This system is installed on the Kyiv University high-performance computing cluster.
   
4. SSI (Single System Image) http://bjbrew.org/cpq/ssic_linux/index.htm. Distributed operating system for Linux as analog of UnixWare clusters. The support of single process area, distributed file system, process migration, distributed memory, providing of reliability and in the case of nodes failure its recovery is in progress. Open source code.
   
5. BEOWULF. http://www.scyld.com/, http://www.beowulf.org/. One of the first projects on creating distributed operating system on the base of Linux. Actually it is a set of computers with single administration. It can be used as distributed computational system on the base of MPI/PVM. Open source code. It is enough stable.
   
6. GFS (Global File System) http://sistina.com/gfs/ . Global file system for Linux. Parallel maintenance of several SCSI devices and network hard drives is supported. Real time modification of file system size, distributed blocking manager, system of failure recovery are also supported. Open source code. It is unstable in network.
   
7. DRBD http://www.complang.tuwien.ac.at/reisner/drbd/. http://www.complang.tuwien.ac.at/reisner/drbd/. Network RAID-1 ( Network mirroring of hard disk partitions). Alfa version. Open source code.
   
8. DLM (Distriguted Lock Manager) http://oss.software.ibm.com/developerworks/projects/dlm/. System of distributed blocking for Linux. Alfa version. Open source code.
   

Systems of cluster monitoring and providing of reliability:

1. VA Cluster Manager http://vacm.sourceforge.net/. High level system that enables cluster nodes monitoring. In the case of failure it performs hard system reboot with power down. It is necessary appropriate hardware support. It is stable. Open source code.


2. Kimberlite http://oss.missioncriticallinux.com/projects/kimberlite/. Failure recovery system for Linux. Open source code.


3. FailSafe http://oss.sgi.com/projects/failsafe/ . System of monitoring and failure recovery for Linux. Open source code.


4. Piranha http://ha.redhat.com/ . System of network load distribution and failure recovery for Linux. Open source code. Beta version.


5. Linux Virtual Server http://www.linuxvirtualserver.org/. System of load distribution, scaling and failure recovery for network servers. It is stable. Open source code.

Parallel processing systems:

1. MPICH http://www-unix.mcs.anl.gov/mpi/mpich/ MPI (Message Passing Interface) implementation http://kiev.parallel.ru/tech/tech_dev/mpi.html . Detailed information is on http://kiev.parallel.ru .


2. LAM http://www.lam-mpi.org/ MPI (Message Passing Interface) implementation.


3. PVM (Parallel virtual machine) http://www.epm.ornl.gov/pvm/pvm_home.html. High level system that enables usage of heterogeneous computer systems for parallel processing.


4. Pov Ray (The Persistence of Vision Raytracer) http://www.povray.org/. System of 3D rendering and animation. MPI is supported. http://www.verrall.demon.co.uk/mpipov/ PVM support http://www-mddsp.enel.ucalgary.ca/People/adilger/povray/pvmpov.html.


5. Radiator. http://www.cs.cmu.edu/~ajw/software/ . System of ray tracing with parallel processing.