Program Level Control of Network Delay for Parallel Asynchronous Iterative Applications
P J Joseph and Sriram Vajapeyam
IISc-CSA-TR-1996-7
(October 1996) Available formats: [ps] [ps.gz]
Software distributed shared memory (DSM) platforms on networks of
workstations tolerate large network latencies by employing one of several
weak memory consistency models. Fully asynchronous parallel iterative
algorithms oer an additional degree of freedom to tolerate network
latency: they do not synchronize shared references and behave correctly
when supplied outdated shared data. However, these algorithms have a high
communication to computation ratio and can flood the network with messages
in the presence of large delays. We propose a method of controlling
asynchronous iterative methods wherein the reader of a shared datum imposes
an upper bound on its age via use of a blocking Global Read primitive. This
reduces the overall number of iterations executed by the reader, thus
controlling the amount of shared updates generated. Experiments for a fully
asynchronous linear equation solver running on a network of 10 IBM RS/6000
workstations show that the proposed Global Read primitive provides
significant performance improvement.
Please bookmark this technical report as http://aditya.csa.iisc.ernet.in/TR/1996/7/.Problems ? Contact techrep@csa.iisc.ernet.in
[Updated at 2009-10-22T06:42Z]