Mardi Gras Conference 2008 Invited Speakers

To Distribute or Not to Distribute, That is the Question in Petascale and Beyond

While there is general consensus that computing platforms underlying the grid infrastructures will continue to evolve, variance in the speed of technology acceleration in HPC is causing many of the assumptions made in the early days of grid to no longer hold. Such divergence in the metrics, as well as wider proliferation of related technologies such as Web2.0 and Clouds, will be changing the optimal design of the overall grid infrastructure towards more centralization of resources, much like the modern-day Internet with its two tier-structure of clients versus the datacenters. Based on our recent experiences with our TSUBAME supercomputer, which is currently Japan's fastest machine according to the Top500, and its next petascale generation design thereof, we will discuss the future design of multi-petascale grids with such machines being constituent massive resource nodes instead of vast distribution, as well as more advanced notions of computing where distribution of compute resources might be fundamental.

GRID computing in the Geospatial Web: The role of standards

For millennia, philosophers and academicians have understood that location and the properties of location - or geography - dictate the form and function of human activity. Every human activity happens somewhere and "somewhen". More recently, scientists and researches have increasing relied on geospatial information, including observations from static and dynamic sensing systems, to better understand and model the complex global dynamics of the earth's atmosphere and oceans. The volume of location based content collected is increasing exponentially. Given the environmental and political issues facing humanity, the ability to discover, access, filter, and integrate geospatial data using standard interfaces and content models in support of simulation and modeling systems is more critical than ever. At the same time, there is a tremendous need for the computing power necessary to properly use and reuse these data in decision support and modeling applications. It is at this juncture where the work of the OGC community on geospatial standards and the work of the GRID community intersect. This presentation will discuss the issues and potential solutions for the integration of geospatial content and services into the GRID infrastructure to create a geospatially enabled GRID. Topics such as federated catalogues and registries, content interoperability, the integration of real time sensor systems, and authentication will be addressed.

2008 - The Year of Global Telepresence

The concept of Telepresence is at least fifty years old, being quite pervasive in science fiction of the 1950s and 1960s. By the late 1980s prototypes using commercial telecommunications were being carried out by research labs in industry and universities, several of which I was involved with. Today, the California Institute for Telecommunications and Information Technology, a UCSD/UCI partnership, has a variety of projects underway exploring persistent 1-10 gigabit/s optical paths connecting people and devices on local, regional, national, and global scales. We are also developing large scale visualization walls, termed "OptIPortals," containing tens to hundreds of millions of pixels, which create large "pixel real estate" for remote collaboration. As part of our digital cinema project, CineGrid, we are experimenting with using four thousand line resolution (4k) video streams carried over dedicated gigabit/sec optical light paths to establish a Telepresence on a global scale. In 2008, many national and international sites will link up their OptIPortals over 1 or 10 gigabit/s light paths, with embedded HD or digital cinema streams, creating a global-scale collaboratory.