Open Source Linux Distributed Computing Software

A Java middleware API supporting broker-less, type+topic-based publish/subscribe and request/reply messaging for both intra- and inter-application, object-level communication. Broker-less: messages are transmitted directly between eBus applications. There is no message broker (i.e. server) in between. Type+topic addressing: eBus clients subscribe to a message class (type) and message topic, providing stronger typing than just topic-based subscriptions. Object-level communication: eBus delivers messages directly to client objects, not to a client process.

FLoM is a free open source distributed lock manager that can be used to synchronize shell commands, scripts and custom developed software. When used in a shell environment, FLoM manages process synchronization in the same way that "nice" manages process prioritization. Shell commands, scripts and custom programs can be synchronized inside a single system or in a network of IP connected systems. FLoM supports different types of abstract resources: simple, numeric, set, hierarchical and transactional sequence. libflom, the library shipped with FLoM, allows you to develop your own application using a simple C, C++, Java, Perl, PHP or Python client API. SSL/TLS security for network communication, X.509 certificates for peer to peer mutual authentication.

OpenHMS brands a collection of projects and many subprojects developed by Health Market Science. This is a robust collection of development libraries, APIs, and tools oriented around data manipulation and professional software development.

CLara is a framework that enables you to access the computing power of many graphics processors in an IP-based network using a kind of Client/Proxy/Server model. Its programming interface conforms to the OpenCL 1.0 standard.

Cluster Networks are a new style of neural simulation / neural network modeling, that models networks of neural populations ("clusters") that transform and transmit information using precisely-timed, graded bursts ("pulses" or "volleys") of firing.

This is an ALPHA version - the RELEASE version is now on github - https://github.com/fimtra/datafission.

JAva DIstributed Framework provides a secure and flexible Framework for distributed grid and volunteer computing applications (separate client & server Framework). For more informations take a look at the project page http://jadif.sourceforge.net

Mpaxs provides a simple abstraction for executing Callables and Runnables either locally within the same Virtual Machine, on the same host via RMI, or on remote machines, also via RMI. It supports launching of new compute hosts via an extensible API, currently featuring local compute host launching and grid compute host launching using the DRMAA API.

C Library to manage a pool of event/task in a persistent way to assure that your events/tasks won't be deleted because of a failure. Events/tasks are saved on a FS. if FS is NFS, NFS availability checks are made. (comes with a python binding)

PyMW is a Python module for parallel master-worker computing in a variety of environments. With the PyMW module, users can write a single program that scales from multicore machines to global computing platforms.

Roomy is a programming language extension for writing parallel disk-based applications. All details of parallelism and disk I/O are hidden within the Roomy library.

Soutei is a trust-management system for access control in distributed systems. Soutei policies and credentials are written in a declarative logic-based language. Soutei policies are modular, concise, readable, supporting conditional delegation.

Tiny Distributed Computing Library is an easy to use cross-platform library for performing distributed computations. It is written in C++ and designed with good performance and modularity in mind.

Java Remote Method Call provides a Framework for remote method invocation on Java objects which solves the limitations of Java RMI. It provides a way for bidirectional communication of object states, between Client and Server.

LIXA (LIbre XA) is an open source and free Transaction Manager implementing the distributed transaction processing "XA specification" and "TX (transaction demarcation) specification" according to the X/Open CAE Specification. LIXA implements even XTA: XA Transaction API, an innovative API that implements XA transactional context passing among different applications. LIXA is a Transaction Manager but it's not a Transaction Monitor: this is the distinguishing feature of the project. LIXA technology enables every application container, like a web server or a shell, to become a two phase commit application server. The client/server architecture of LIXA allows many application containers to share a single LIXA (state) server: this is ideal when horizontal scalability is a must and many identical application containers must refer to a single transactional environment. LIXA can be used with the C, C++, Java, Python and COBOL programming languages.

A parallel-programming framework for concurrently running large numbers of small autonomous jobs, or microthreads, across multiple cores in a CPU or CPUs in a cluster. NeuraNEP emulates a distributed processing environment capable of handling millions of microthreads in parallel, for example running neural networks with millions of spiking cells. Microthreads are general processing elements that can also represent non-neural elements, such as cell populations, extracellular space, emulating sensory activity, etc. NeuraNEP handles microthread scheduling, synchronization, distribution and communication. This project is a fork of SpikeOS (sourceforge.net/projects/spikeos) and represents a major update to that code base, including a scripting interface and low-level rewrite of several components. SpikeOS was oriented towards computational modeling. NeuraNEP is oriented toward neural network research.

proJMS is a library for Processing that allows the easy usage of JMS. It allows to publish and consume JMS messages and to create peer to peer communication between processing applications in a network. Refer to the wiki for installation and usage instructions.

Slimgrid is a Java library for grid computations which is lighter than other ones (JPPF, Hadoop, ...). The main design goals are: minimalism, simplicity, pervasiveness. If you need to grab something which does not require you to comprehend massive and complex API's, do exhaustive configurations and installations, is robust and reliable, uses just one port for all management and communication, then SlimGrid is the right choice. The SlimGrid is built on top of the Apache's ZooKeeper library for coordination of distributed systems and Apache Curator awesome API simplification. Slimgrid allows also one-way synchronization of grid application folders (from the server to workers). This is useful when the grid app has been updated. Slimgrid is very robust. One of his successful mission was to perform more than 140.000 tasks in the grid with 40+ computers located in three different countries (Norway, Germany, Canada) and behind various firewalls (University labs).