DocumentCode
2384262
Title
Near Overhead-free Heterogeneous Thread-migration
Author
Veldema, Ronald ; Philippsen, Michael
Author_Institution
Dept. of Comput. Sci., Erlangen-Nuremberg Univ., Erlangen
fYear
2005
fDate
Sept. 2005
Firstpage
1
Lastpage
10
Abstract
Thread migration moves a single call-stack to another machine to improve either load balancing or locality. Current approaches for checkpointing and thread migration are either not heterogeneous or they introduce large runtime overhead. In general, previous approaches add overhead by instrumenting each function in a program. The instrumentation costs are then even incurred when no thread migration is performed. In this respect our system is near-overhead free: nearly no overhead is caused if no migration is performed. Our implementation instead generates meta-functions for each location in the code where a function is called. These functions portably save and rebuild activation records to and from a machine-independent format. Each variable of an activation record is described in terms of its usages in a machine-independent `usage descriptor string´ to enable heterogeneous, near overhead free thread migration with as few as possible changes to a compiler. Our resulting thread migration solution is, for example, able to move a thread between an x86 machine (few registers, 32 bits) and an Itanium machine (many registers, 64 bits). Furthermore, we (optionally) move the decision on when and where to migrate to the application programmer instead of implementing a fixed ´fits-all´ heuristics as in previous approaches
Keywords
checkpointing; resource allocation; Itanium machine; checkpointing; large runtime overhead; load balancing; load locality; machine-independent format; near overhead-free heterogeneous thread-migration; single call-stack; thread migration; usage descriptor string; x86 machine; Checkpointing; Computer science; Instruments; Load management; Packaging machines; Program processors; Programming profession; Registers; Runtime; Yarn;
fLanguage
English
Publisher
ieee
Conference_Titel
Cluster Computing, 2005. IEEE International
Conference_Location
Burlington, MA
ISSN
1552-5244
Print_ISBN
0-7803-9486-0
Electronic_ISBN
1552-5244
Type
conf
DOI
10.1109/CLUSTR.2005.347042
Filename
4154085
Link To Document