Download Latest Version
concx-2.3.1.zip (2.6 MB)
Get an email when there's a new version of Avian Parallel Computing
| Name | Modified | Size | Downloads / Week |
|---|---|---|---|
| ConcX2-3 archive | 2018-07-28 | ||
| ConcX2-2 archive | 2017-12-20 | ||
| ConcX2-1 archive | 2017-02-02 | ||
| JavaOne Presentation | 2011-10-11 | ||
| concx-2.3.1.zip | 2018-07-28 | 2.6 MB | |
| concx-2.3JavaDocs.zip | 2017-12-23 | 1.3 MB | |
| README.TXT | 2017-12-20 | 2.3 kB | |
| Getting Started With Avian Computing 2-3a.pdf | 2017-12-20 | 6.7 MB | |
| Totals: 8 Items | 10.7 MB | 0 |
The Concurrency Explorer (ConcX) version 2.3 is now available. This latest version offers a bunch of improvements, including most importantly, the ability to run from a jar file in a Linux terminal window or a Windows cmd window. Just download the zip file and extract its contents to any convenient folder and run ConcX from there with the following command:
java -jar concx.jar Nimbus
It's quick and easy because all the lib resources are included in the zip file, as are all the flock and data files. And the whole zip file is less than 3 MB. Get the zip file from this page or from the aviancomputing.net blog site.
The source code can be downloaded from the Code page either as a "Code Snapshot" zip file or as a Subversion checkout. Be sure to also download "Getting Started with Avian Computing" guide for details of how to install and run ConcX.
Note that ConcX is built using Java 8 and NetBeans 8.x. There is very little in the code that depends on Java 8 so it could probably be retrofitted to Java 7 if really needed. And NetBeans is used because of the excellent (and free) GUI designer it makes available standard.
For those of you who are eager to learn more about Avian Computing, please point your browser to aviancomputing.net where you will find a full discussion of why Avian Computing exists and the problems that it attempts to solve.
If you want the short version of "Why ConcX", Avian Computing tries to make parallel programs quicker and easier to develop. It pursues that goal by providing:
1) a nature-based model to make it easier to think about parallel programs
2) a framework that minimizes coding that is also rich in diagnostic features
3) a framework that simplifies (all but eliminates) parallel coding hassles
4) automatic logging of events for each thread with individually selected logging levels
5) dynamic workload sensing and can autoajust number of threads to fit the workload
6) interactive starting and stopping of individual threads without crashing the running app
7) a Linda-like parallel information sharing system that borrows from JavaSpaces
8) highly de-coupled objects that can have their operation sequences rearranged without code changes
Enjoy!