Please note: This seminar will take place in DC 1304.
Yuanhao
Wei,
Postdoctoral
Researcher
Computer
Science
and
Artificial
Intelligence
Laboratory,
MIT
Concurrent programming is becoming increasingly important as systems are scaling up by increasing the number of processors rather than the speed of a single processor. However, concurrent programming can be very difficult and error-prone.
This talk will cover several simple, general, and efficient techniques for designing concurrent algorithms and data structures. This includes techniques for lock-freedom, for taking consistent snapshots of shared memory, and for safe memory management. With the right abstractions and algorithms, these general techniques can be very efficient in practice, achieving competitive performance with the fastest hand-optimized data structures, while also providing strong theoretical guarantees.
Bio: Yuanhao Wei is a Postdoctoral Associate at the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL). Previously, he obtained his PhD from CMU advised by Professor Guy Blelloch.
His research interests are in developing efficient algorithms and easy-to-use abstractions for multi-core machines. His work in this area has been recognized by a best paper award from the ACM Symposium on Principles and Practice of Parallel Programming (PPoPP’22).