Distributed and parallel computing have been investigated for many years but recently research on this topic has gained new impetus due to the explosive growth of the Internet, the occurrence of cluster and Grid computing infrastructures, and the availability of portable programming languages.
The distributed and parallel systems group is developing a programming environment and a tool set for Cluster and Grid computing. In the context of our Grid research we are designing and implementing
- new programming paradigms and methods
- application development and submission systems
- intelligent middleware services including
- scheduling
- resource management
- performance instrumentation, measurement, analysis
- performance prediction
- networking
Substantial advances in the field of programming languages and methods enable the programmer to write effective programs at a machine-independent level. However, as parallelization and optimization of programs is far from being automated, there is a clear need for useful, efficient and accurate tools to support this process. The following tools are being developed within the framework of ASKALON, a programming environment and tool set for cluster and grid computing:
- AKSUM: an automatic performance bottleneck analysis tool
- Performance PROPHET: a performance modeling and prediction system
- SCALEA: A performance instrumentation, measurement, and analysis tool for parallel programs.
- SCALEA-G: A Unified Performance Monitoring and Analysis System for Grids.
- ZENTURIO: an automatic performance experiment management system
- Integrated Graphical User Interface: a Java-based coordination and visualization system for tools
Java is very popular due to its ability to support code mobility, object-orientation, portability, multi-threading, synchronization mechanisms, and communication APIs. Although there is no agreement whether Java can ever satisfy the needs of high-performance driven applications, it is commonly used to exploit medium grain shared memory parallelism based on threads as well as medium to coarse grain data and task parallelism based on remote method invocation and socket communication. Most ASKALON tools so far support mostly Fortran90 (including MPI, OpenMP, HPF, mixed OpenMP/MPI, and mixed HPF/OpenMP) based distributed and parallel programs. However, we have also developed a novel Java-based programming paradigm for performance-oriented parallel and distributed computing, called JavaSymphony which will be supported by the ASKALON tools in the near future.
For more information contact research(at)dps.uibk.ac.at.