Synergy User Manual and Tutorial. - THE CORE MEMORY
Synergy User Manual and Tutorial. - THE CORE MEMORY
Synergy User Manual and Tutorial. - THE CORE MEMORY
You also want an ePaper? Increase the reach of your titles
YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.
<strong>Synergy</strong> <strong>User</strong> <strong>Manual</strong> <strong>and</strong> <strong>Tutorial</strong><br />
Introduction<br />
Red text: Copied <strong>and</strong> pasted from syng_man.ps by Dr. Shi<br />
The emergence of low cost, high performance uni-processors forces the enlargement of<br />
processing grains in all multi-processor systems. Consequently, individual parallel<br />
programs have increased in length <strong>and</strong> complexities. However, like reliability, parallel<br />
processing of any multiple communicating sequential programs is not really a functional<br />
requirement.<br />
Separating pure functional programming concerns from parallel processing <strong>and</strong> resource<br />
management concerns can greatly simplify the conventional ``parallel programming''<br />
asks. For example, the use of dataflow principles can facilitate automatic task<br />
scheduling. Smart tools can automate resource management. As long as the application<br />
dependent parallel structure is uncovered properly, we can even automatically assign<br />
processors to parallel programs in all cases.<br />
<strong>Synergy</strong> V3.0 is an implementation of above ideas. It supports parallel processing using<br />
multiple ``Unix computers'' mounted on multiple file systems (or clusters) using TCP/IP.<br />
It allows parallel processing of any application using mixed languages, including parallel<br />
programming languages. <strong>Synergy</strong> may be thought of as a successor to Linda 1 , PVM 2 <strong>and</strong><br />
Express 3 .<br />
Our need to store <strong>and</strong> process data has been continually increasing for thous<strong>and</strong>s of years.<br />
This need has lead to the development of complex storage, communication, numerical<br />
<strong>and</strong> processing systems. The information in this section was wholly obtained from<br />
sources freely available on the Internet, which are cited in the references section. Much<br />
of it was obtained from timelines, encyclopedias <strong>and</strong> academic Web pages. The accuracy<br />
of information collected from the Internet was checked by using multiple corroborating<br />
resources <strong>and</strong> eliminating contradictory information.<br />
1 Linda is a tuple space parallel programming system lead by Dr. David Gelenter, Yale University. Its<br />
commercial version is distributed by the Scientific Computing Associates, New Heaven, NH.<br />
2 PVM is a message passing parallel programming system by Oak Ridge National Laboratory, University<br />
of Tennessee <strong>and</strong> Emory University.<br />
3 Express is a commercial message passing parallel programming system by ParaSoft, CA.<br />
5