This title appears in the Scientific Report :
2012
OSPRI: An Optimized One-Sided Communication Runtime for Leadership-Class Machines
OSPRI: An Optimized One-Sided Communication Runtime for Leadership-Class Machines
Abstract—Partitioned Global Address Space (PGAS) programming models provide a convenient approach to implementing complex scienti?c applications by providing access to a large, globally accessible address space. This paper describes the design, implementation and performance of a new one-sided commu...
Saved in:
Personal Name(s): | Hammond, Jeff R. (Corresponding author) |
---|---|
Dinan, James / Balaji, Pavan / Kabadshow, Ivo / Potluri, Sreeram / Tipparaju, Vinod | |
Contributing Institute: |
Jülich Supercomputing Center; JSC |
Imprint: |
2012
|
Physical Description: |
10 p. |
Conference: | The 6th Conference on Partitioned Global Address Space Programming Models, Santa Barbara, California (USA), 2012-10-10 - 2012-10-12 |
Document Type: |
Contribution to a conference proceedings |
Research Program: |
Fast Multipole Method Computational Science and Mathematical Methods |
Publikationsportal JuSER |
Abstract—Partitioned Global Address Space (PGAS) programming models provide a convenient approach to implementing complex scienti?c applications by providing access to a large, globally accessible address space. This paper describes the design, implementation and performance of a new one-sided communication library that attempts to meet the needs of PGAS models, particularly Global Arrays, but hopefully also PGAS languages like UPC and CAF. In this work, we describe a new communication runtime for PGAS models such as GA, termed OSPRI (One-Sided PRImitives). OSPRI presents several changes in architecture from conventional one-sided communication systems that make it better suited for emerging leadersip class machines. We describe the implementation of the the IBM Blue Gene/P target for OSPRI and demonstrate signi?cant improvements in latency, bandwidth, and scalability over tuned ARMCI and GA implementations on this system. The performance and scalablity of this library validate the design choices and should provide useful insight for implementers of related communication middleware. |