CLuMPI (OpenCL under MPI) is an implementation of the OpenCL interface which uses MPI as a back end
the bbsc - BinBase Cluster System is a specification for the distributed calculation of jobs using j2ee and a cluster. We provide an implementation for Rocks Linux Cluster (Sun Grid Engine). This project has been merged into the BinBase system a couple of years back.
The OpenSSI project is a comprehensive clustering solution offering a full, highly available SSI environment for Linux. Goals for OpenSSI Clusters include availability, scalability and manageability, built from standard servers.
Alchemi is a .NET grid computing framework that allows you to painlessly aggregate the computing power of intranet and Internet-connected machines into a virtual supercomputer (computational grid) and to develop applications to run on the grid.
Ridiculously simple, Prevayler provides transparent ACID persistence and replication for plain old Java objects. It is literally several thousand times faster than using a relational database through JDBC.
Construct and management virtual clusters which the nodes of the clusters running on the virtual machine rather than native machine. Fast install and switch virtual clusters. Efficiently storage vm images. support variant VMMs.
Maui Scheduler is an advanced reservation HPC parallel batch scheduler for use with Linux and BSD clusters. Maui provides a complete scientific scheduling solution, supporting running custom parallel and MPI jobs over Myrinet and ethernet.
The Beowulf Distributed Process Space (BProc) is a set of Linux kernel modifications for easy process management and remote process creation in Beowulf style clusters.
Programming libraries for writing an XMLDC (XML Distributed Computing) applications in C++. XMLDC is a XML-based protocol for remote procedure calls (RPC) between different systems. XMLDC is more powerful than XML-RPC and much simpler than SOAP or CORBA.
A Virtual Domains manager for Xen powered by SmartFrog. Description, deployment and coordinated lifecycle management of pools of remote Xen virtual domains.
Distributed RDF Repository based on peer-to-peer network overlay
[Ckpt Object-oriented Abstraction Layer] COAL is an object-based abstraction layer utilizing the SAForum's AIS CKPT (Checkpoint) specification to provide HA access to cluster-wide data encapsulated within local, user-defined objects.
MICO/mt plans to document and implement multi-threading for MICO, RTCORBA, and AMI interfaces
Open Distributed Locking Manager
Elastica is a highly efficient and extensible, rules-based load-balancer for JBoss that adds dynamic behavior to EJB load balancing. Rules can be defined to redirect EJB requests according to request data, server performance data, or even the time of day
High performance distributed in-memory key/value store
Infinispan is an open source, Java based data grid platform. ***IMPORTANT*** Starting with Infinispan 5.0.0.FINAL, Infinispan releases are no longer hosted in Sourceforge. They can now be located in www.jboss.org/infinispan/downloads
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.
The Gridbus Grid Service Broker, is a service-oriented, economy and data-aware resource broker system that uses a simple, extensible middleware independent architecture that mediates access to distributed resources on a local / Global Grid of computers.
A JDBC driver proxy that adds light-weight, transparent, fault tolerant clustering capability to any underlying JDBC driver.
The Parallel Compositing library enables developers to write scalable 3D graphics applications. The library does sort-last compositing, and works on both clusters and SMP machines. The API aims to be the MPI of parallel compositing.
A data parallel scientific programming model. Compiles efficiently to different platforms like distributed memory (MPI), shared memory multi-processor (pthreads), Cell BE processor, Nvidia Cuda, SIMD vectorization (SSE, Altivec), and sequential C++ code.
The Imperative Skeleton Library (iskel) is a project aimed towards higher-level abstraction of parallel programming. Algorithmic Skeletons are being provided as library functions which can then be used in parallel programming.
BSPonMPI is an implementation of the BSPlib standard on top of MPI. Both MPI and BSPlib are API's of communication routines meant for parallel computing, but BSPlib is easier to learn and its performance easier to predict.
The project Mandala helps the development of concurrent and/or distributed applications. It is based on the asynchronous reference concept which provide asynchronous and potentially remote method invocation.
A set of utility programs to work with the spread messaging system.