The Shared Scientific Toolbox is a library that facilitates development of efficient, modular, and robust scientific/distributed computing applications in Java. It features multidimensional arrays with extensive linear algebra and FFT support, an asynchronous, scalable networking layer, and advanced class loading, message passing, and statistics packages.
jmemcached is a fast network available cache daemon. It is protocol-compatible with memcached, but written in Java and suitable for applications with portability concerns, where Java is the preferred solution, or for using the memcached protocol in embedded applications with alternate storage engines. Existing clients for memcache work unmodified. It can run as a standalone daemon or be embedded inside an existing Java application.
Dapper, or "Distributed and Parallel Program Execution Runtime", is a tool for taming the complexities of developing for large-scale cloud and grid computing, enabling the user to create distributed computations from the essentials: the code that will execute, along with a dataflow graph description. It supports rich execution semantics, carefree deployment, a robust control protocol, modification of the dataflow graph at runtime, and an intuitive user interface.
The overall objective of the XtreemOS project is the design, implementation, evaluation, and distribution of a grid operating system (called XtreemOS) with native support for virtual organizations (VO). XtreemOS is capable of running on a wide range of underlying platforms, from clusters to mobiles. It is based on Mandriva Linux, with support to come for other distributions later.
chunkd is a simple storage service based on the GET, PUT, and DELETE operations. It is meant to be used in building larger, replicated, distributed storage systems. Clients connect via TCP and remotely manage storage. Typical applications include distributed filesystems and other distributed storage applications.
DAC (Dynamic Agent Computations) is a novel software framework designed for implementing multi-agent systems that describe parallel computations. The whole system is easy to configure and extend, but also very efficient and scalable. Moreover, the technology that is used (JMS, Cajo, JMX) ensures high reliability of the framework, which can be used in a production environment.
Java distributed framework is a framework for distributed grid and / or volunteer computing. It's divided into a server and client library. You can create new or implement it into existing applications in no time; you don't need knowledge about network connections, sockets, etc. The Framework does almost everything automatically. It provides secure automatic client <-> server communications, unique IDs, automatic resending of jobs to new clients if needed, user stats, and much more. The client framework supports the detection of the computer's user state (idling, away, online, etc.). It also offers many other useful features and helpers for developing a distributed client application.
Makeflow is a workflow engine for executing large complex applications on clusters, clouds, and grids. It can be used to drive several different distributed computing systems, including Condor, SGE, and the included Work Queue system. It does not require a distributed filesystem, so you can use it to harness whatever collection of machines you have available. It is typically used for scaling up data-intensive scientific applications to hundreds or thousands of cores.