Exploiting Object Locality in JavaParty, a Distributed Computing Environment for Workstation Clusters
نویسندگان
چکیده
In a distributed programming environment with location transparency, fast access to remote resources is absolutely critical for eÆcient program execution but it is not suÆcient. Locality optimization will try to group objects according to their communication patterns and replace remote access by local access whenever possible. Locality optimization is based on the assumption that local access always is much faster than remote access. Using conventional communication packages like RMI, this assumption is void. A remote object always needs an access layer around it to allow access to the object by means other than direct reference. Transparent access to a remote object also needs to overcome this access layer, even in a local access. Direct reference access can not be used, because of di erent parameter passing semantics. In this paper we present techniques to provide transparent local access to potentially remote objects in the same magnitude as direct Java access.
منابع مشابه
JavaParty - Transparent Remote Objects in Java
Java’s threads offer appropriate means either for parallel programming of SMPs or as target constructs when compiling add-on features (e.g. forall constructs, automatic parallelization, etc.) Unfortunately, Java does not provide elegant and straightforward mechanisms for parallel programming on distributed memory machines, like clusters of workstations. JavaParty transparently adds remote objec...
متن کاملA Dynamic Task Graph Parallelization Approach
The general purpose Finite Element system PERMAS [1] has been parallelized by a dynamic task graph approach capable to support shared and distributed parallel computer ar-chitectures as well as workstation clusters. The method used to parallelize this large application software package is of high generality and has the capability to parallelize all mathematical operations in a FEM analysis – no...
متن کاملParallel Computing Environments and Methods
Recent advances in high-speed networks, rapid improvements in microprocessor design, and availability of highly performing clustering software implementations enables cost-effective high-performance parallel computing on clustered low cost workstations and PCs. Such clusters are very attractive because they rely on available of the shelf hardware and software technologies. To simpha programming...
متن کاملExploiting Temporal & Spatial Constraints on Distributed Shared Objects∗
Gigabit network technologies have made it possible to combine workstations into a distributed, massivelyparallel computer system. Middleware, such as distributed shared objects (DSO), attempts to improve programmability of such systems, by providing globally accessible ’object’ abstractions. Researchers have developed consistency protocols for replicated ’memory’ objects. These protocols are we...
متن کاملA Distributed Object Based Framework for Parallel Computations
The computational and compositional features are very important while constructing parallel software for the workstation clusters. However, lack of suitable supporting environment for parallel software development makes most existing distributed parallel software systems very weak in these two aspects, especially in the compositional feature. In this paper, a distributed object based framework ...
متن کامل