SHARCNET
Multicomputer Systems r A multicomputer system comprises of a number of independent machines linked by an interconnection network. r Each computer executes its own program which may access its local memory and may send and receive messages over the network. r The nature of the interconnection network has been a major topic of research for both academia and industry. r Distributed systems is one type of multicomputer system. What about others?
Multiprocessor Usage Scientific and engineering applications often require loops over large vectors e.g., matrix elements or points in a grid or 3D mesh. Applications include Computational fluid dynamics Scheduling (airline) Health and biological modeling Economics and financial modelling (e.g., option pricing)
Multiprocessor Usage r It should be noted that people have been developing “clusters” of machines that are connected using Ethernet for parallel applications. r The first such cluster (developed by two researchers at NASA) had machines and was connected using 10 Mb Ethernet. r This is known as the Beowulf approach to developing a parallel computing and the clusters are sometimes called Beowulf clusters.
Sharcnet r UWO has taken a leading role in Canada in exploiting the concepts behind the Beowulf cluster. r High performance clusters: “Beowulf on steroids” m Powerful “off the shelf” computational elements m Advanced communications r Geographical separation (local use) r Connect clusters: emerging optical communications r This is referred to as Shared Hierarchical Academic Research Computing Network or Sharcnet
Sharcnet r One cluster is called “Great White” r Processors: m 4 alpha processors: 833Mhz (4p-SMP) m 4 Gb of memory m 38 SMPs: a total of 152 processors r Communications m 1 Gb/sec ethernet m 1.6 Gb/sec quadrics connection r November 2001: #183 in the world m Fastest academic computer in Canada m 6 th fastest academic computer in North America
Sharcnet Great White (in Western Science Building)
Sharcnet r Extend “Beowulf” approach to clusters of high performance clusters r Connect clusters: “clusters of clusters” m Build on emerging optical communications m Initial configuration used optical equipment from telecommunications industry r Collectively a supercomputer!
Sharcnet GUELPH MAC UWO Optical communication Clusters across Universities (initial cluster)
Sharcnet r In 2004, UWO received an investment of 56 million dollars from the government and private industry (HP) to expand Sharcnet. r With the new capabilities, Sharcnet could be in the top 100 or 150 of supercomputers. r Will be the fastest supercomputer of its kind – I.e.,a distributed system where nodes are clusters.
Sharcnet
Sharcnet Applications r Applications running on Sharcnet come from all sorts of domains including m Chemistry m Bioinformatics m Economics m Astrophysics m Material Science and Engineering