This work is joint with Rabin Patra, Hari Balakrishnan and Eric Brewer.

Bio: Ramakrishna Gummadi is a post-doc at the MIT Computer Science and Artificial Intelligence Laboratoray (CSAIL). He obtained his B.Tech. from IIT-Madras, M.S. from UC Berkeley and Ph.D. from USC, all in Computer Science. His dissertation was about reliable and efficient programming languages for sensor networks. He is interested in building scalable and reliable systems and networks based on sound principles. His awards include a UC Regents Graduate Fellowship, a best paper awarded out of all 2001 Journal of Computer Networks papers, a best poster/demo award at SenSys 2004, and an award at the ACM Student Research Competition (SRC) held at PLDI 2007.

This talk will say why we think we can write a new OS, and why we think we should. Hand-in-hand with increasing hardware parallelism is increasing hardware diversity, even within a single machine. Furthermore, the drive towards multicore programmability is beginning to result in interesting language and runtime features whose I/O and scheduling requirements may not be well served by existing OS structure.

Barrelfish seeks to meet these challenges by viewing a multicore machine more as a networked system than as a single, monolithic computer, and applying results from distributed computing to scaling a single OS instance across many heterogeneous cores. We also apply knowledge representation techniques to allow the OS and applications to reason about the richness of the hardware at runtime, in the interests of continuous performance optimization. I'll talk about how these approaches lead to a novel way of structuring an OS, and the current status and future directions of the system. ork).ABSTRACT: The Barrelfish OS is a new open-source operating system for heterogeneous multicore systems being developed at ETH Zurich, in conjunction with Microsoft Research in Cambridge. This talk will say why we think we can write a new OS, and why we think we should. Hand-in-hand with increasing hardware parallelism is increasing hardware diversity, even within a single machine. Furthermore, the drive towards multicore programmability is beginning to result in interesting language and runtime features whose I/O and scheduling requirements may not be well served by existing OS structure. Barrelfish seeks to meet these challenges by viewing a multicore machine more as a networked system than as a single, monolithic computer, and applying results from distributed computing to scaling a single OS instance across many heterogeneous cores. We also apply knowledge representation techniques to allow the OS and applications to reason about the richness of the hardware at runtime, in the interests of continuous performance optimization. I'll talk about how these approaches lead to a novel way of structuring an OS, and the current status and future directions of the system.

Please see this link for more information.

Thursday, April 10, 2008
4:00-5:00pm
2nd Floor, Tea Room
CS Building

Refreshments will be served.

See this page for more information.

Please see this link for more information.