Also much discussion concerning power management–but not clear how serious anyone is about the topic. (E.g., no real discussion of what tradeoffs will be made to reduce power consumption, or mention of genuinely low-power systems, such as SiCortex.)
He reports discussion on clouds
Considerable discussion of “clouds” (whatever they may be) much of it naive and ill-informed. ascinating to hear the same expansive expectations expressed (without any apparent doubts) for “clouds” as we heard five years ago from the most fullsome proponents off grids.
And some apparently good remarks as well
Perhaps the most thoughtful remarks on these topics were from Ignacio Llorente, who is working with relevant technologies via his Globus GridWay and OpenNebula projects. He characterized clouds as “a paradigm for the on-demand provision of virtualized resources as a service” (correctly identifying virtualization rather than interfaces as the key advance) and grids as “the technology that will allow for cloud interoperability.”
My own thoughts on what people are calling “clouds” might be important in HPC for providing remote hosting of HPC services for organizations that can no longer afford (or invest the time it takes) to build the enormous and enormously expensive infrastructures it takes to host the big beasties. There are other possible advantages as well, including the possibility of designing services with “zero” downtime (from a combination of over provisioning and server/process virtualization) as well as good economies of scale for purchase. For smaller organizations these hosted HPC services could provide an important surge processing requirement in situations that require a short burst of HPC for problems that are usually run in house. One area of potentially large benefit for these organizations is the possibility of software licensing (a substantial expense for some of the commercial packages) “as you need it.”
More, including some discussion on the European Grid Initiative, in the post.