Simon See

As most of us will have already know that Niagara take on the school that we should have simple pipeline and with multicore and multithread. The first Niagara processor has only one float-point unit shared among all the cores and threads. In Niagara 2, the number of floating units is increased to 1 per core which make is attractive and interesting for some of the floating point computation intensive application. While the chip’s peak floating point performance of 11,2 GFlop/s quite moderate, the high potential of the Niagara 2 is revealed, when many threads are active and the high memory bandwidth of some 60 GB/s (theoretically) can be exploited - a frequent bottleneck of standard architectures when executing technical applications.

Take at look at the result at http://www.rz.rwth-aachen.de/ca/k/raw/?lang=en

One of my main research interest in in using Economic or Market Theory in Grid resource allocation. We started some work in the late 2005. Below are some of the paper which we have published. A Market-Based Framework for Trading Grid Resources, Melvin Koh, Jie Song, Liang Peng & Simon See, The 3rd International Workshop on Grid Economics and Business Models - GECON 2006, Singapore, 16 May 2006. Survey on the Application of Economic and Market Theory for Grid Computing, Xie Xia, Huang Jin, Wu Song, Jin Hai, Melvin Koh Chee Kian, Song Jie & Simon See, Technical Report APSTC-TR-2006-005, 30 August 2006. A Proposed Simple Grid Market Framework - Part I, Jie Song, Melvin Koh, Liang Peng & Simon See, Technical Report APSTC-TR-2005-001, 8 August 2005. In 2006, we were invited to participate in a EU-FP project call SORMA. SORMA deals with the development of methods and tools for an efficient market-based allocation of resources, through a self-organizing resource management system, using market-driven models supported by extensions for Grid infrastructures. Topics addressed include open Grid markets, economics-driven Grid middleware, and intelligent tools. Brokerage of resources over the Grid will benefit Grid consumers, resource owners, outsourcing providers, and new intermediaries. http://www.iw.uni-karlsruhe.de/sormang/

Supercomputing is just round the corner. This time is going to be held in Reno. Not sure if there is a correlation between gambling and high performance computing. Anyway, it remind me of my first supercomputing conference when I got out of the university. At that time working as a research scientist in a national lab in Singapore. Being trained as a numerical analyst, I was assigned to develop simulation programs for all sort of systems. Since than I've moved from ISIM (a simulation consulting company) to SGI and then to Sun. I remember it was March of 2001. I was a devoted member of the SGI then and with much persuasion, Marc Hamilton (currently VP for Solaris Marketing) recruited me into Sun. I was asked to build HPC for Sun in Asia Pacific. It was then we build Asia Pacific Science and Technology Center (http://apstc.sun.com.sg). Deepak, Melvin and a whole lots of young engineers helped to get it started. Over the period of 5 years, APSTC has done some interesting projects, published 50+ papers and have lots of collaboration with varies universities. It is a challenge but a interesting journey. Recently Sun has make tremendous in road into the HPC segment. With the new X64 (AMD and Intel) based systems, Sun has put herself back on the HPC map. I'm fortunate to be asked to run the Global HPC Solution team. This is part of the growth target and I'm very excited about work we are entrusted with. I also hold position in both universities (National University of Singapore and Nanyang Technological Univ) as associate research professors. This is a very combination as it allow me to funnel some of the industry research work to my students which Sun does not have time or the resources to do. It is fun and I'm looking forward to have many good fun years in Sun.