Global Parallel Computation, Networks and Geometry

Add to your list(s) Download to your calendar using vCal

• Peter Braam (Braam Research LLC, USA)
• Tuesday 01 March 2016, 16:00-17:00
• Martin Ryle Seminar Room, Kavli Institute.

If you have a question about this talk, please contact Greg Willatt.

We discuss implementations of distributed algorithms for a broad family of linear algebra problems which lead to perfect load balancing of data movement, storage and computation. The algorithms were likely first discovered by Karmarkar in the 1990’s leveraging finite projective geometries. From this approach, network topologies emerge which exhibit resilience, very low diameters and high cross sectional bandwidth. A few studies of these algorithms in software have shown significant speedup over current practice, subject to proper locality of ingest data. Silicon photonics may allow for effective implementation of the networks. While it has received little attention, this broad mixture of ideas appears to form a viable system architecture for parallel computing, but it also raises questions about implementation methodology and performance studies of parallel algorithms.

This talk is part of the Cavendish Astrophysics Seminars series.

This talk is included in these lists:

• All Cavendish Laboratory Seminars
• All Talks (aka the CURE list)
• Cambridge Centre for Data-Driven Discovery (C2D3)
• Cambridge talks
• Cavendish Astrophysics Seminars
• Centre for Health Leadership and Enterprise
• Chris Davis' list
• Combined External Astrophysics Talks DAMTP
• Cosmology, Astrophysics and General Relativity
• Featured lists
• Interested Talks
• ME Seminar
• Martin Ryle Seminar Room, Kavli Institute
• Neurons, Fake News, DNA and your iPhone: The Mathematics of Information
• School of Physical Sciences
• Thin Film Magnetic Talks
• Trust & Technology Initiative - interesting events
• bld31
• ndk22's list
• ob366-ai4er
• rp587

Note that ex-directory lists are not shown.