Computer Science Thesis Proposal

— 3:30pm

In Person - Reddy Conference Room, Gates Hillman 4405

JATIN ARORA , Ph.D. Student, Computer Science Department, Carnegie Mellon University

Provably Efficient Coscheduling of Computation and Data through Disentanglement

Because of its many desirable properties, such as its ability to control effects and thus potentially disastrous race conditions, functional programming offers a viable approach to programming modern multicore computers. Over the past decade several parallel functional languages, typically based on dialects of ML and Haskell, have been developed. These languages, however, have traditionally underperformed compared to procedural languages (such as C and Java). The primary reason for this is their hunger for memory, which only grows with parallelism, causing traditional memory management techniques to buckle under increased demand for memory. 

Building on a long line of work on parallel memory management, this thesis proposes provably and practically efficient techniques for memory management of parallel functional programs. The key idea behind our techniques is to coschedule the parallel computation with its data, enabling the memory manager to exploit our disentanglement hypothesis—the idea that parallel tasks of a program largely execute independently and avoid side-effecting data that may be accessed by others—for efficiency. We implement these techniques in the MPL compiler for parallel ML and our experimental results show that the techniques can marry the safety benefits of functional programming with performance.

Thesis Committee:

Umut Acar (Chair) 
Guy Blelloch
Robert Harper
Yongshan Ding (Yale University)
Rustan Leino (Amazon)

Additional Information

Add event to Google
Add event to iCal