Guest Speaker: Veljko Milutinovic
DataFlow SuperComputing for BigData Analytics
This presentation analyzes the essence of DataFlow SuperComputing, defines its advantages and sheds light on the related programming model. DataFlow computers, compared to ControlFlow computers, offer speeds up of 20 to 200 (even 2000 for some applications), power reductions of about 20, and size reductions of also about 20. However, the programming paradigm is different, and has to be mastered. The talk explains the paradigm, using Maxeler as an example, and sheds light on the ongoing research in the field. Examples include DataEngineering, DataMining, FinancialAnalytics, etc. A recent study from Tsinghua University in China reveals that, for Shallow Water Weather Forecast, which is a BigData problem, on the 1U level, the Maxeler DataFlow machine is 14 times faster than the Tianhe machine, which is rated #1 on the Top 500 list (based on Linpack, which is a small data benchmark). Given enough time, the talk also gives a tutorial about the programming in space, which is the programming paradigm used for the Maxeler dataflow machines (established in 2014 by Stanford, Imperial, Tsinghua, and the University of Tokyo). Examples (appgallery.maxeler.com and webIDE.maxeler.com).
Prof. Veljko Milutinovic received his PhD from the University of Belgrade (rated #366 world-wide at the U.S. News Report 2015), spent about a decade on various faculty positions in the USA (mostly at Purdue University), and was a co-designer of the DARPAs first GaAs RISC microprocessor. Now he teaches and conducts research at the University of Belgrade, in EE and MATH (Belgrade math is rated #87 world-wide at the U.S. News Report 2015). His research is mostly in datamining algorithms and dataflow computing, with the stress on mappings of algorithms onto architectures. His co-authored paper on matrix multiplication for dataflow received "The IET Premium Award for 2014" (meaning the best paper in IET Computing for 2012 and 2013). His paper on the research methodology was published in an ASCE journal and received a special recognition for the most downloaded paper of the publisher. He published over 20 books with the USA publishers. For 7 of his books, forewords were written by 7 different Nobel Laureates with whom he cooperated on his past industry sponsored projects. He has over 60 IEEE or ACM journal papers, and about 3000 Google Scholar citations.
Host: DK Panda