Name
Abstract | ||
|---|---|---|
Swift/T is a high-level language for writing concise, deterministic scripts that compose serial or parallel codes implemented in lower-level programming models into large-scale parallel applications. It executes using a data-driven task parallel execution model that is capable of orchestrating millions of concurrently executing asynchronous tasks on homogeneous or heterogeneous resources. Producing code that executes efficiently at this scale requires sophisticated compiler transformations: poorly optimized code inhibits scaling with excessive synchronization and communication. We present a comprehensive set of compiler techniques for data-driven task parallelism, including novel compiler optimizations and intermediate representations. We report application benchmark studies, including unbalanced tree search and simulated annealing, and demonstrate that our techniques greatly reduce communication overhead and enable extreme scalability, distributing up to 612 million dynamically load balanced tasks per second at scales of up to 262,144 cores without explicit parallelism, synchronization, or load balancing in application code. |
Year | DOI | Venue |
|---|---|---|
2014 | 10.1109/SC.2014.30 | New Orleans, LA |
Keywords | Field | DocType |
concurrency control,optimising compilers,parallel programming,simulated annealing,tree searching,Swift/T,application benchmark,application code,asynchronous tasks,code optimization,communication overhead reduction,compiler optimizations,compiler transformations,data-driven task parallel execution model,data-driven task parallelism,deterministic scripts,heterogeneous resource,high-level language,homogeneous resource,intermediate representations,load balancing,lower-level programming models,parallel applications,parallel codes,scalable implicit task parallelism,serial codes,simulated annealing,unbalanced tree search | Functional compiler,Implicit parallelism,Task parallelism,Explicit parallelism,Computer science,Compiler correctness,Parallel computing,Compiler,Optimizing compiler,Data parallelism,Distributed computing | Conference |
ISSN | Citations | PageRank |
2167-4329 | 15 | 0.74 |
References | Authors | |
28 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Timothy G. Armstrong | 1 | 373 | 21.73 |
| Justin M. Wozniak | 2 | 464 | 35.32 |
| Michael Wilde | 3 | 15 | 0.74 |
| Foster Ian | 4 | 22938 | 2663.24 |