Recent accelerators such as GPUs achieve better
cost-performance and watt-performance ratio, while the
range of their application is more limited than general
CPUs. Thus heterogeneous clusters and supercomputers
equipped both with acclerators and general CPUs
are becoming popular, such as LANL RoadRunner and
our own TSUBAME supercomputer. Under the assumption
that many applications will run both on CPUs
and accelerators but with varying speed and power consumption
characteristics, we propose a task scheduling
scheme that optimize overall energy consumption
of the system. We model task scheduling in terms of
the scheduling makespan and energy to be consumed
for each scheduling decision. We define acceleration
factor to normalize the effect of acceleration per each
task. The proposed scheme attempts to improve the energy
efficiency by effectively self adjusting the threshold
on the scheduling using the acceleration factor depending
on the results of its own scheduling. Although in
the paper we adopted the popular EDP (Energy-Delay
Product) as the optimization metric, our scheme is agnostic
on the optimization function. Simulation studies
on various sets of tasks with mixeed acceleration factors,
the overall makespan closely matched the theoretical
optimal, while the energy consumption was saved
up to 13.8%.
各種検索
サポート
ヘルプ