JPPF is a computational grid framework for Java focused on performance and ease of use. It provides a set of tools and APIs to enable the parallelization of CPU intensive applications, and distribute their execution over a network of heterogenous nodes. It features platform independence thanks to Java 1.5, does not require you to deploy your application classes to a server, scales up to millions of nodes, has a built-in fail-over mechanism on all the framework's components, and has a monitoring and administration GUI tool to enable remote monitoring of the server health and server shutdown/restart operations.
| Tags | Scientific/Engineering Software Development Libraries Java Libraries Clustering/Distributed Networks |
|---|---|
| Licenses | LGPL Apache 2.0 |
| Operating Systems | Mac OS X Windows Windows POSIX Linux Unix |
| Implementation | Java |
| Translations | English |
Recent releases


Changes: Submitted tasks can now fetch data from the client while executing. A node's configuration can now be updated remotely without restart. A new sample illustrates how tasks can communicate with each other. Numerous bugs have been fixed, improving the overall stability and performance.


Changes: A new API enables users to run non-Java processes on the nodes. JPPF is now integrated with GigaSpaces XAP. Jobs and tasks can now be prioritized. The configuration of a client connection pool is now easy and intuitive.


Changes: Servers are now automatically detected, removing the need to use configuration files. The administration console was enhanced for greater usability. Administration commands can now be sent to multiple nodes at once. The nodes' CPU usage can now be managed by setting the priority of the execution threads. Critical class loading issues were fixed in the J2EE JCA connector.


Changes: Tasks can now be defined from plain old Java objects. A new JPPF Quick Start Guide is now availabe online and offline. A new management feature enables resetting a node's task counter. Improvements to the remote JMX connectivity facilitate the JPPF administration through firewalls. Bugs were fixed in the peer-to-peer communication between servers.


Changes: Tasks can now be defined from existing code using annotations. A new, simplified API for defining and submitting tasks was implemented. The amount of logging on disconnection events from the server was reduced to a manageable level. Multiple bugs were fixed that prevented the nodes from running as services. Multiple fixes were made to the JPPF screensaver.
- All comments
Recent commentsRe: what is its relationship to clustering (e.g. JBoss)?
Hello Jamie,
Thank you for your comments and question.
Indeed, JPPF is not a cluster, but rather a computational grid framework.
The generally agreed upon difference between clusters and grids is that clusters are made of a set of tightly coupled, homogeneous systems (hardware/OS/software) in a single location or complex, whereas grids are loosely coupled, heterogeneous and widely distributed in nature.
In a larger definition, a grid provides access to computational and storage resources across organizational boundaries. In effect, it virtualizes these resources from the user's perspective, as the user doesn't need to know what these resources are made of or where they are located to use them.
A J2EE application server such as JBoss provides clustering capabilities. Its main goal is to serve a large number of relatively short-lived interactions (or transactions) with a large number of users.
On the other hand, a computational grid will serve requests for long-lived intensive computations that can take hours or days or even longer to complete.
In short, frameworks like JBoss and JPPF provide capabilities in different areas, where they do not compete with each other. It is my belief that they instead complement each other. To demonstrate this, we (JPPF team) have developed a connector between JPPF and the major J2EE application servers in the market, including JBoss.
I invite you to find more about it on the JPPF web site at http://www.jppf.org
I hope this answers your question,
-Laurent
> This project sounds very interesting. My
> question is: what is its relationship
> with clustering setup for example in
> JBoss? Is it a similar thing but in a
> new development? or is it complete a new
> technology?
what is its relationship to clustering (e.g. JBoss)?
This project sounds very interesting. My question is: what is its relationship with clustering setup for example in JBoss? Is it a similar thing but in a new development? or is it complete a new technology?