Abstract
This paper discusses mechanisms to exploit static information to increase locality in the Flagship fine-grained parallel machine. The machine supports a global address space across its distributed physical store. Load balancing is entirely dynamic and global-to-local address mapping to achieve high performance is carried out in a number of ways. A program is compiled into a graph of packets. This paper considers how static information is used to enable the dynamic mapping of the graph of packets to the distributed stores to improve performance and increase locality. Simulation figures suggest utilising static information results in some access taking less than half the time.
Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
