| High speed network and improved microprocessor performance ard making networks of computers an appealing vehicle for parallel computing. By relying solely on commodity hardware and software, network of computers can offer cost-effective high performance and high availability computing. This new wave in high performance computing is popularly called as Cluster Computing.Today Cluster Computing has become an extremely effective way to solve many problems in the large-scale scientific computing and engineering field. However, there are many factors to affect the efficiency of cluster computing, such as task gratuity, processors allocating and network topology, in which load balancing and task-scheduling strategy is a crucial one and thus has become a hot research spot in the field of parallel processing.Parallel Virtual Machine (PVM) is a software package that permits a heterogeneous collection of 'Cluster Computers' to function as a single high-performance parallel machine. Thus large computational problems can be solved more cost effectively by using the aggregate power and memory of many computers. PVM is portable and runs on a wide variety of modern platforms. With tens of thousands of users, PVM has become the defacto standard for distributed computing world-wide.In this paper, firstly, that is analyzed which is the systemic composition of cluster network parallel computing, message passing environment: PVM, parallel programming of PVM on Win32 platform, scheduling strategy of load balancing and the general model of scheduling problem and stimulatingly the factors affecting the scheduling efficiency are summarized. For these different kinds of factors, some new strategies are promoted and some improved metheds are provided: in order to resolve the bottle necklace problems in centralized task scheduling strategy and simultaneously obtain the better load balancing capability, these ideas of two-tier scheduling and active reporting are presented; by introducing the conception of valve... |
PDF Full Text Request