Salim Hariri, Manish Parashar, JongBaek Park, Fang-Kuo Yu, and Geoffrey Fox, "A Message Passing Interface for Parallel and Distributed Computing", To be presented at the 2 nd International Symposium on High Performance Distributed Computing, Spokane, Washington, July 1993.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....(1) implementing the protocol in a special communication processor, thus offloading the host from protocol processing, and (2) making the protocol implement the basic functionality of message passing libraries. The communication processor (HIP) is designed to offer rates close to the medium speed [8]. The protocol includes separate procedures for transmitting short and long messages, as well as simple error and flow control functions in order to ensure reliability. The protocol provides applications with HPDC specific services: point to point communications, group communications, process ....

S. Hariri, J. Park, F.-K. Yu, M. Parashar, and G. C. Fox. A MessagePassing Interface for Parallel and Distributed Computing. In Proceedings of the 2nd International Symposium on High Performance Distributed Computing, pages 84--91, 1993.

....(1) implementing the protocol in a special communication processor, thus offloading the host from protocol processing, and (2) making the protocol implement the basic functionality of messagepassing libraries. The communication processor (HIP) is designed to offer rates close to the medium speed [8]. The protocol includes separate procedures for transmitting short and long messages, as well as simple error and flow control functions in order to ensure reliability. The protocol provides applications with HPDC specific services: point to point communications, group communications, process ....

S. Hariri, JB Park, F-K Yu, M. Parashar, and G. C. Fox. A Message Passing Interface for Parallel and Distributed Computing. In Proceedings of the 2nd International Symposium on High Performance Distributed Computing, pages 84--91, 1993.

....most ATM environments, HPDC applications will share the ATM network with usual applications. In order to efficiently manage both types of applications, the architecture of the ATM network has to provide adequate support for both of them. For this purpose, we have adopted the architectural view in [8], where the network is logically divided into two subnetworks, one for usual applications and the other for HPDC applications. The former can continue to be based on usual protocol stacks like the TCP IP protocol suite, while the latter has to rely on a low latency reliable protocol ....

S. Hariri, J. Park, F.-K. Yu, M. Parashar, and G. C. Fox. A Message Passing Interface for Parallel and Distributed Computing. In Proceedings of the 2nd International Symposium on High Performance Distributed Computing, pages 84--91, 1993.

....of workstation clusters to achieve supercomputer like performance. We consider that, in ATM networks, HPDC applications will run in conjunction with usual LAN applications. The architecture of the ATM network has to account for this requirement, thus we have adopted the architectural view in [4], where the network is logically divided into two subnetworks, one for usual LAN applications and the other for HPDC applications. The former can be based on usual protocol stacks like the TCP IP protocol suite, while the latter can rely on a low latency protocol architecture. The ....

S. Hariri, JB Park, F-K Yu, M. Parashar, and G. C. Fox. A Message Passing Interface for Parallel and Distributed Computing. In Proceedings of the 2nd IEEE International Symposium on High Performance Distributed Computing, pages 84--91, 1993.

....most ATM environments, HPDC applications will share the ATM network with usual applications. In order to efficiently manage both types of applications, the architecture of the ATM network has to provide adequate support for both of them. For this purpose, we have adopted the architectural view in [8], where the network is logically divided into two subnetworks, one for usual applications and the other for HPDC applications. The former can continue to be based on usual protocol stacks like the TCP IP protocol suite, while the latter has to rely on a lowlatency reliable protocol architecture. ....

S. Hariri, J. Park, F.-K. Yu, M. Parashar, and G. C. Fox. A MessagePassing Interface for Parallel and Distributed Computing. In Proceedings of the 2nd International Symposium on High Performance Distributed Computing, pages 84--91, 1993.

....magnitude and diversity would require a general, cost effective, scalable, yet powerful computing model which will be able to efficiently support its varied computational and communication requirement. It is this realization that has spurred intense research in heterogeneous computing environments [2, 3, 4, 5, 6, 1, 7, 8]. We believe that the future of parallel computing lies in the integration of the plethora of specialized architectures into a single Heterogeneous High Performance Computing (HHPC) environment that allows them to cooperate in solving complex problems (Figure 1) The HHPC environment will ....

Salim Hariri, Manish Parashar, JongBaek Park, Fang-Kuo Yu, and Geoffrey Fox, "A Message Passing Interface for Parallel and Distributed Computing", To be presented at the 2 nd International Symposium on High Performance Distributed Computing, Spokane, Washington, July 1993.

....magnitude and diversity would require a general, cost effective, scalable, yet powerful computing model which will be able to efficiently support its varied computational and communication requirement. It is this realization that has spurred intense research in heterogeneous computing environments [2, 3, 4, 5, 6, 1, 7, 8]. We believe that the future of parallel computing lies in the integration of the plethora of specialized architectures into a single Heterogeneous High Performance Computing (HHPC) environment that allows them to cooperate in solving complex problems (Figure 1) The HHPC environment will ....

Salim Hariri, Manish Parashar, JongBaek Park, Fang-Kuo Yu, and Geoffrey Fox, "A Message Passing Interface for Parallel and Distributed Computing", To be presented at the 2 nd International Symposium on High Performance Distributed Computing, Spokane, Washington, July 1993.

....magnitude and diversity would require a general, cost effective, scalable, yet powerful computing model which will be able to efficiently support its varied computational and communication requirement. It is this realization that has spurred intense research in heterogeneous computing environments [2, 3, 4, 5, 6, 1, 7, 8]. We believe that the future of parallel computing lies in the integration of the plethora of specialized architectures into a single Heterogeneous High Performance Computing (HHPC) environment that allows them to cooperate in solving complex problems. The HHPC environment will capitalize on ....

Salim Hariri, Manish Parashar, JongBaek Park, Fang-Kuo Yu, and Geoffrey Fox, "A Message Passing Interface for Parallel and Distributed Computing", To be presented at the 2 nd International Symposium on High Performance Distributed Computing, Spokane, Washington, July 1993.

....completion of the request N 3 : Notifies application Events: Figure 8: Application to application latency example, the two RTUs can be configured to transmit and or receive data over high speed channels while the TEU is transferring data to from the host. More detail description can be found in [1, 2, 3]. 3 Performance Analysis In this section, we analyze the application to application latency in HLAN and show its effective transfer rates under different network conditions. We analyze performance of D net for point to point communication and S net for broadcast communication. We develop both ....

S. Hariri, J. Park, F. Yu, and M. Parashar, and G. C. Fox, " A Message Passing Interface for Parallel and Distributed Computing," Proceedings of the 2nd International Symposium on High-Performance Distributed Computing (HPDC-2), IEEE, pp. 84--91, July 1993.

....and parallelize the application at this level will be based on emerging standards (High Performance Fortran [4] High Performance C [1] etc. Support for architecture independent communication between tasks will be provided via standard communication libraries (based on standards such as MPI [5, 14]) and the semantics of object oriented method invocation. As the task graph is annotated in the various layers of the SDM, the user will be able to record information about the nature of the application that will be useful to the compilation and runtime managers. These hints will allow the ....

Hariri, S., Park, J., Yu, F., Parashar, M., and Fox, G., A Message Passing Interface for Parallel and Distributed Computing, Proceedings of the Second International Symposium on High Performance Distributed Computing, pp. 84-91, July 1993.

.... PICL send0 recv0 message0 recvinfo0 probe0 PVM ISIS iPSC snd rcv vsnd vrcv rcvinfo rcvmulti probe probemulti [x]bcast[2] x]bcast l[2] reply reply l Group Communication exbroadcast excombine exconcat bcast0 g[op]0 [1] csend crecv isend irecv hsend hrecv csendrecv isendrecv g[type] op] [3] gopf gcol Synchronization exsync exsem sync0 barrier0 barrier ready waituntil token tool waitall waitone gsync Configuration Control Management exopen exclose exgrid exparam exload exstart abort fmulti fsingle open0 close0 load0 setarc0 getarc0 initiate leave pstatus terminate isis init ....

....isis rexec msg read msg write getcube relcube load cread cwrite iread iwrite Exception Handling pvm perror monitor tool watch tool handler (not addressed ) not addressed ) 1] For PICL, op= max, min, sum, prod, comb, and, or, xor . 2] For ISIS, x = a, c, f, g, m is the type of broadcast. [3] For iPSC, type = d(double) i(integer) s(real) op = high, low, sum, prod, and, or, xor . Table 1: A characterization of message passing primitives for parallel and distributed computing a Transfer Engine Unit (TEU) a crossbar switch, and two Receive Transmit units (RTU 1, RTU 2) The ....

S. Hariri, J. B. Park, F-K. Yu, and M. Parashar, and G. C. Fox, " A Message Passing Interface for Parallel and Distributed Computing," Proceedings of the 2nd International Symposium on HighPerformance Distributed Computing (HPDC-2), IEEE, pp. 84--91, July 1993.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC