Efficient Parallel Computing on Workstation Clusters
نویسندگان
چکیده
We present novel hardand software that efficiently implements communication primitives for parallel execution on workstation clusters. We provide low communication latencies, minimal protocol, zero operating system overhead, and high throughput. With this technology, it is possible to build effective parallel systems using off-the-shelf workstations. Our goal is to develop a standard interfaceboard and the necessary software for interfacing any number of computers, from a workstation to a cabinet full of workstation-boards.
منابع مشابه
Approaches to Support Parallel Programming on Workstation Clusters: A Survey
The goal of this report is to survey state of the art and existing approaches for parallel programming on workstation clusters with special emphasis on object{oriented programming. First, workstation clusters as parallel computing platforms are characterized and fundamental concepts for parallel programming are discussed. Then, an overview of existing tools, systems, languages, and environments...
متن کاملA new architecture for efficient parallel computing in workstation clusters: conceptions and experiences
The acceptance of parallel computing in workstation clusters has increased in the past years. One important reason for this is the cost-efficiency of workstation clusters as an alternative to specialized distributed-memory parallel computer systems. A potential bottleneck for distributedmemory architectures is the interconnection network between the processing elements. This is the main disadva...
متن کاملParallel Algorithm for Filamentation of High-Power Super-Short Laser Pulses
An efficient parallel algorithm is proposed for solving the filamentaion problem of high-power femtosecond lasers pulses in gases. It was demonstrated that the algorithm possesses the property of coarse-grained parallelism and has a good scale-up characteristics when used on workstation clusters.
متن کاملLoad Balancing for Parallel Loops in Workstation Clusters
Load imbalance is a serious impediment to achieving good performance in parallel processing. Global load balancing schemes cannot adequately manage to balance parallel tasks generated from a single application. Dynamic loop scheduling methods are known to be useful in balancing parallel loops on shared-memory multiprocessor machines. However, their centralized nature causes a bottleneck even fo...
متن کاملcient Parallel Computing on Workstation Clusters
We present novel hardand software that e ciently implements communication primitives for parallel execution on workstation clusters. We provide low communication latencies, minimal protocol, zero operating system overhead, and high throughput. With this technology, it is possible to build e ective parallel systems using o -the-shelf workstations. Our goal is to develop a standard interfaceboard...
متن کامل