S. Hariri, J. Park, M. Parashar, and G. C. Fox. A Communication System for High-Performance Distributed Computing. Concurrency, Practice and Experience. Special Issue: High Performance Distributed Computing, June 1994.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....is oriented to provide an efficient API, like the Active Messages proposal described in [12] where messages are integrated into the ongoing computation while avoiding redundant buffering operations. Other approaches build parallel computing specific transport protocols, like the proposal in [6] which implements the most common functions of current message passing libraries. Finally, some mechanisms allow to carry out computations while blocked for communications provided that the expected data are not required. In [7] this is achieved by means of parallel computing specific ....

S. Hariri, J. Park, M. Parashar, and G. C. Fox. A Communication System for High-Performance Distributed Computing. Concurrency, Practice and Experience. Special Issue: High Performance Distributed Computing, June 1994.

....Several authors have proposed mechanisms to improve some of these issues. We classify these proposals in three groups, according to the primary issue the enhancement impacts on: 1) the ATM API; 2) the transport protocol; and (3) the application context. Examples of these approaches are [17] [7], and [20, 9] respectively. 4.1. API level enhancement mechanisms In Section 3, the implementation of the API adopted for the experiments has shown to be an important bottleneck. Therefore, the research on alternative APIs becomes a key 5 requirement to achieve enhanced performance for ATMbased ....

....unreliable, therefore they were designed to be very general in order to handle complex failure scenarios. These characteristics have lead to complicated and therefore time consuming protocol implementations. The high speed and high reliability of current networks allows for simpler protocols. In [7], a communication system specifically tailored for HPDC is presented. The architecture of this system is shown in Figure 3(a) In this environment, the transport protocol is called HCP. Its high speed is achieved by (1) implementing the protocol in a special communication processor, thus ....

S. Hariri, J. Park, M. Parashar, and G. C. Fox. A Communication System for High-Performance Distributed Computing. Concurrency, Practice and Experience. Special Issue: High Performance Distributed Computing, June 1994.

....Several authors have proposed mechanisms to improve some of these issues. We classify these proposals in three groups, according to the primary issue the enhancement impacts on: 1) the ATM API ; 2) the transport protocol ; and (3) the application context . Examples of these approaches are [16] [7], and [19, 9] respectively. 4.1 API level enhancement mechanisms In [16] a multicomputer communication mechanism, namely Active Messages is adapted to ATM. This mechanism is designed to implement a very fast, low granularity RPC model implementation [17] Each transmitted message includes a ....

....unreliable, therefore they were designed to be very general in order to handle complex failure scenarios. These characteristics have lead to complicated and therefore time consuming protocol implementations. The high speed and high reliability of current networks allows for simpler protocols. In [7], a communication system specifically tailored for HPDC is presented. The architecture of this system is shown in Figure 3(a) In this environment, the transport protocol is called HCP. Its high speed is achieved by (1) implementing the protocol in a special communication processor, thus ....

S. Hariri, JB Park, M. Parashar, and G. C. Fox. A Communication System for High-Performance Distributed Computing. Concurrency, Practice and Experience. Special Issue: High Performance Distributed Computing, June 1994.

....This mechanism is an adaption to ATM of the Active Messages proposal for multicomputers. Unfortunately, the interface is somewhat complex for the programmer since it involves some hard restrictions, specially in multithreaded environments. At a transport level, a suitable approach is shown in [23], where a transport protocol specially designed for supporting parallel computing is proposed. This protocol implements the most common functions found in messagepassing libraries. In addition, the protocol is conceived to be partially implemented in the host interface, thus offloading the CPU of ....

....in Figure 5. As discussed in [24, 25] the architecture considers three levels, and the mechanisms addressed to supporting parallel computing are included in one or several of these levels. The idea of a specific architecture for parallel computing coexisting with other applications appears in [23], although that proposal is oriented to supporting a particular mechanism. In contrast, the proposal discussed in the work is intended to integrate whatever mechanisms are specifically conceived for parallel computing. The three levels conforming this architecture are the following: level ....

S. Hariri, JB Park, M. Parashar, and G. C. Fox. A Communication System for High-Performance Distributed Computing. Concurrency, Practice and Experience. Special Issue: High Performance Distributed Computing, June 1994.

....[10] where messages are integrated into the ongoing computation as soon as received, and most of operation is carried out in user space, thus avoiding redundant buffer copying and buffering. Other mechanisms construct parallel computing specific transport protocols. For instance, the proposal in [11] implements the most common functions of current message passing libraries and is intended to be implemented in a dedicated communications processor. Finally, some mechanisms take advantage of the idle time induced by communications to carry out computations not requiring the expected data. In ....

S. Hariri, JB Park, M. Parashar, and G. C. Fox. A Communication System for High-Performance Distributed Computing. Concurrency, Practice and Experience. Special Issue: High Performance Distributed Computing, June 1994.

No context found.

S. Hariri, J.B. Park, M. Parashar, G.C. Fox. Communication system for highperformance distributedcomputing. Concurrency: Practice and Experience 6,4 (June 1994), 251--270.

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