Abstract — Global communication costs in future single-chip multiprocessors will increase linearly with distance. In this pa-per, we revisit the issues of locality and load balance in order to take advantage of these new costs. We present a technique which simultaneously migrates data and threads based on vectors spec-ifying locality and resource usage. This technique improves per-formance on applications with distinguishable locality and imbal-anced resource usage. 64 % of the ideal reduction in execution time was achieved on an application with these traits while no improve-ment was obtained on a balanced application with little locality. I
Thread migration is established as a mechanism for achieving dynamic load sharing and data lo-cality...
While the paradigm offered by SMP designs is a relatively clean one, programming paradigms offered o...
Moving threads is a theoretically interesting approach for mapping the computation of an application...
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Comp...
Multi-core chips allow thread and program level parallelism thus increasing performance. How-ever, t...
We present a completely new kind of approach for mapping the computation of an application to MP-SOC...
Chip-multiprocessors (CMPs) have become the mainstream chip design in recent years; for scalability ...
Thread migration moves a single call-stack to another machine to improve either load balancing or lo...
Thread migration is established as a mechanism for achieving dynamic load sharing and data locality....
Thread migration is established as a mechanism for achieving dynamic load sharing and data locality....
This dissertation presents a new methodology for developing parallel distributed programs in a serie...
Networks of workstations are fast becoming the standard envi-ronment for parallel applications. Howe...
The page migration problem occurs in managing a globally addressed shared memory in a multiprocessor...
As the number of cores integrated on a single chip continues to increase, communication has the pote...
The transition to multi-core architectures can be attributed mainly to fundamental limitations in cl...
Thread migration is established as a mechanism for achieving dynamic load sharing and data lo-cality...
While the paradigm offered by SMP designs is a relatively clean one, programming paradigms offered o...
Moving threads is a theoretically interesting approach for mapping the computation of an application...
Thesis: Ph. D., Massachusetts Institute of Technology, Department of Electrical Engineering and Comp...
Multi-core chips allow thread and program level parallelism thus increasing performance. How-ever, t...
We present a completely new kind of approach for mapping the computation of an application to MP-SOC...
Chip-multiprocessors (CMPs) have become the mainstream chip design in recent years; for scalability ...
Thread migration moves a single call-stack to another machine to improve either load balancing or lo...
Thread migration is established as a mechanism for achieving dynamic load sharing and data locality....
Thread migration is established as a mechanism for achieving dynamic load sharing and data locality....
This dissertation presents a new methodology for developing parallel distributed programs in a serie...
Networks of workstations are fast becoming the standard envi-ronment for parallel applications. Howe...
The page migration problem occurs in managing a globally addressed shared memory in a multiprocessor...
As the number of cores integrated on a single chip continues to increase, communication has the pote...
The transition to multi-core architectures can be attributed mainly to fundamental limitations in cl...
Thread migration is established as a mechanism for achieving dynamic load sharing and data lo-cality...
While the paradigm offered by SMP designs is a relatively clean one, programming paradigms offered o...
Moving threads is a theoretically interesting approach for mapping the computation of an application...