ParADE
OpenMP-based parallel programming environment for SMP cluster systems.About
Currently, commodity off-the-shelf microprocessors and network components are widely used as building blocks for parallel computers. This trend has made cluster systems consisting of symmetric multiprocessors (SMPs) attractive platforms for high-performance computing. Even though it is easy to configure cluster systems, specifically of small scale, it is challenging to utilize them easily and fully.
To achieve the goal of easy and high-performance programming, we propose a parallel programming environment for SMP clusters. The ParADEsystem provides a unified programming model of OpenMP while the runtime system executes a program in a hybrid of message passing and shared address space.
The key features are as follows.
1. Intelligent OpenMP translator
The basic role of the translator is to convert the OpenMP program into a hybrid form of MPI and software DSM interfaces.
2. Explicit message passing primitives
3. Multi-threaded software distributed shared memory (SDSM)
4. Home-based lazy release consistency (HLRC) with migratory home
5. Dynamic load balancing
6. Adaptive computing