Title :
SCPlib: a concurrent programming library for programming heterogeneous networks of computers
Author :
Watts, Jerrell ; Taylor, Stephen ; Nilpanich, Sirikunya
Author_Institution :
Scalable Concurrent Programming Lab., Syracuse Univ., NY, USA
Abstract :
This paper describes the Scalable Concurrent Programming Library (SCPlib), basic technology that supports irregular applications on scalable concurrent hardware and heterogeneous computing environments. The library is optimized to take advantage of the best available underlying communication and synchronization on a variety of high-performance multicomputers, shared-memory multiprocessors, and networked PCs and workstations. It also provides a framework for heterogeneous communication and file I/O, load balancing, and dynamic granularity control. The effectiveness of the library has been demonstrated on a variety of industrial strength applications
Keywords :
computer networks; multiprocessing systems; parallel programming; resource allocation; software libraries; synchronisation; SCPlib; Scalable Concurrent Programming Library; dynamic granularity control; file input output; heterogeneous communication; heterogeneous computer networks; high-performance multicomputers; load balancing; networked personal computers; scalable concurrent hardware; shared-memory multiprocessors; synchronization; workstations; Application software; Communication system control; Computer networks; Concurrent computing; Hardware; Libraries; Load management; Plasma simulation; Workstations; Yarn;
Conference_Titel :
Information Technology Conference, 1998. IEEE
Conference_Location :
Syracuse, NY
Print_ISBN :
0-7803-9914-5
DOI :
10.1109/IT.1998.713403