DocumentCode
2422563
Title
High Productivity Languages for Parallel Programming Compared to MPI
Author
Spetka, Scott ; Hadzimujic, Haris ; Peek, Stephen ; Flynn, Christopher
Author_Institution
Comput. & Inf. Sci., SUNY Inst. of Technol., Utica, NY
fYear
2008
fDate
14-17 July 2008
Firstpage
413
Lastpage
417
Abstract
Two new high productivity parallel programming languages, Chapel (Cray) and X10 (IBM), are compared to the message passing interface (MPI) parallel programming environment. The study qualitatively compares language features used to develop a prototype publication/subscription system. Our results show that the global partitioned address space (GPAS) and support for other parallel processing abstractions, employed by each of the new languages, increases productivity and reduces time to solution for parallel program development.
Keywords
message passing; middleware; parallel languages; parallel programming; Chapel language; X10 language; global partitioned address space; high productivity language; message passing interface; parallel programming; prototype publication-subscription system; Concurrent computing; Costs; High performance computing; Information management; Message passing; Parallel programming; Productivity; Programming profession; Prototypes; Subscriptions;
fLanguage
English
Publisher
ieee
Conference_Titel
DoD HPCMP Users Group Conference, 2008. DOD HPCMP UGC
Conference_Location
Seattle, WA
Print_ISBN
978-1-4244-3323-0
Type
conf
DOI
10.1109/DoD.HPCMP.UGC.2008.41
Filename
4755902
Link To Document