Orthogonal Processor Groups for Message-Passing Programs
نویسندگان
چکیده
We consider a generalization of the SPMD programming model to orthogonal processor groups. In this model different partitions of the processors into disjoint processor groups can be exploited in a single parallel implementation. The parallel programming model is appropriate for grid based applications with computations and dependence pattern mainly aligned in horizontal or vertical directions. For those applications we propose a systematic development process for message-passing programs using orthogonal processor groups. The development process starts with an SPMD-like specification of tasks indicating horizontal and vertical sections. A mapping to orthogonal processor groups realizes a group SPMD execution model with different orthogonal processor groups and a final transformation generates the corresponding MPI program. In this paper, we describe the approach and present example applications which can benefit from an implementation with orthogonal processor groups. The runtimes are measured on a Cray T3E.
منابع مشابه
The Impact of Communication Style on Machine Resource Usage for the iWarp Parallel Processor
Programs executing on a private-memory parallel system exchange data by explicitly sending and receiving messages. Two communication styles have been identified for such systems: memory communication (each message exchanged between two processors is buffered in memory, e.g. as in message passing) and systolic communication (each word of a message is transmitted directly from the sender processo...
متن کاملImproving the Performance of Message-Passing Applications by Multithreading
Achieving maximum performance in message-passing programs requires that calculation and communication be overlapped. However, the program transformations required to achieve this overlap are error-prone and add significant complexity to the application program. We argue that calculation/communication overlap can be achieved easily and consistently by executing multiple threads of control on eac...
متن کاملPerformance and modularity benefits of message-driven execution
Processor idling due to communication delays and load imbalances are among the major factors that affect the performance of parallel programs. Need to optimize performance often forces programmers to sacrifice modularity. This paper focuses on the performance benefits of message-driven execution, particularly for large parallel programs composed of multiple libraries and modules. We examine mes...
متن کاملAn Improved DSM System Design and Implementation
In this paper, an Improved Distributed Shared Memory (IDSM) system, a hybrid version of shared memory and message passing version is proposed. This version effectively uses the benefits of shared memory in terms of ease of programming and message passing in terms of efficiency. Further it is designed to effectively utilize the stateof-art multicore based network of workstations and supports sta...
متن کاملCRANIUM: An Interface for Message Passing on Adaptive Packet Routing Networks
Cranium is a processor-network interface for an intercon-nection network based on adaptive packet routing. Adaptive networks relax the restriction that packet order is preserved; packets may be delivered to their destinations in an arbitrary sequence. Cranium uses two mechanisms: an automatic-receive interface for packet serialization and high performance, and a processor-initiated interface fo...
متن کامل