With uniprocessors running out of steam, parallel architectures provide a realistic path towards scalable performance. Nevertheless, shared-memory multiprocessors are limited by the difficulty of writing correct and fast parallel programs. Transactional Coherence and Consistency (TCC) is a new model for shared memory systems that builds upon transactional memory. In TCC, user-defined transactions are the basic unit of parallelism, coherence and consistency, error atomicity, and optimization. TCC simplifies parallel programming by eliminating the need for manual orchestration of parallelism using locks. It also provides a smooth transition from sequential to parallel programs. This talk will introduce the software and hardware aspects of TCC and provide an overview of recent research developments in the TCC project.
Christos Kozyrakis is an Assistant Professor of Electrical Engineering and Computer Science at Stanford University. He holds a B.S. degree from the University of Crete in Greece and a Ph.D. from the University of California at Berkeley. Christos' research focuses on architectures, compilers, and programming models for parallel computer systems. He is currently working on transactional memory techniques that that can greatly simplify parallel programming for the average developer.
More info at:http://csl.stanford.edu/~christos