Showing results for Sciddle 4.0, or, Remote Procedure Calls in PVM.
Search instead for Sciddle 4.0, or, Remote Procedure Calles in PVM.
Sciddle 4.0, or, remote procedure calls in PVM
P Arbenz, W Gander, HP Lüthi, U von Matt - High-Performance Computing …, 1996 - Springer
P Arbenz, W Gander, HP Lüthi, U von Matt
High-Performance Computing and Networking: International Conference and …, 1996•SpringerSciddle 4.0 makes it possible to use remote procedure calls (RPCs) within the PVM
environment. Parallelism is achieved through asynchronous RPCs. No explicit message
passing is necessary. Rather all data transfers occur within the RPCs. In Sciddle, an
application is decomposed into a client process and an arbitrary number of server
processes. Servers are special PVM tasks that are ready to execute RPCs from their clients.
Servers can also start other servers themselves. Thus the topology of a Sciddle application …
environment. Parallelism is achieved through asynchronous RPCs. No explicit message
passing is necessary. Rather all data transfers occur within the RPCs. In Sciddle, an
application is decomposed into a client process and an arbitrary number of server
processes. Servers are special PVM tasks that are ready to execute RPCs from their clients.
Servers can also start other servers themselves. Thus the topology of a Sciddle application …
Abstract
Sciddle 4.0 makes it possible to use remote procedure calls (RPCs) within the PVM environment. Parallelism is achieved through asynchronous RPCs. No explicit message passing is necessary. Rather all data transfers occur within the RPCs.
In Sciddle, an application is decomposed into a client process and an arbitrary number of server processes. Servers are special PVM tasks that are ready to execute RPCs from their clients. Servers can also start other servers themselves. Thus the topology of a Sciddle application can be described by a tree structure.
Sciddle applications enjoy the safety and ease of use of RPCs. They are also extremely portable as PVM becomes available on more and more platforms. In this paper we show that the overhead introduced by Sciddle is minimal and can be neglected for applications with large data sets.
Springer
Showing the best result for this search. See all results