Energy-aware data prefetching for multi-speed disks

Abstract

Power consumption of disk based storage systems is becoming an increasingly pressing issue for both commercial and scientific application domains. Prior work proposed several hardware based approaches to reducing disk power consumption by making use of techniques such as spinning down idle disks and rotating them at lower speeds than the maximum speed possible. While such techniques are certainly very important, it is also critical to consider the influence the software can exercise in shaping the power consumption behavior of disk-intensive application programs. Motivated by this, the main goal of this work is to study whether an optimizing compiler can be used for increasing the power benefits that could be obtained from multi-speed disks. Specifically, we propose and experimentally evaluate a compiler-directed energyaware data prefetching scheme for scientific applications that process disk-resident data sets. This scheme automatically determines the prefetch distance for all disk access instructions, the disk speeds to be employed, and the associated disk layouts (striping parameters) in a unified setting. We implemented the proposed approach within an optimizing compiler framework and conducted experiments with several disk-intensive applications. Our experimental evaluation shows that the proposed approach brings significant reductions in disk energy consumption over a state-of-the-art softwarebased I/O prefetching mechanism that does not take into account energy consumption explicitly. Our results also show that the energyaware prefetching scheme does not bring any extra performance penalties and the energy reductions achieved are consistent across a wide spectrum of values of the simulation parameters.

An adaptive energy-conserving strategy for parallel disk systems

“…Recently, several techniques proposed to conserve energy in storage systems include dynamic power management schemes [9], power aware cache management strategies [17], power aware perfecting schemes [18], softwaredirected power management techniques [19], redundancy techniques [19], and multi-speed settings [20]. However, the research on energy-efficient parallel disk systems is still in its infancy.…”

Heat-based dynamic data caching: A load balancing strategy for energy-efficient parallel storage systems with buffer disks

“…Several techniques proposed to conserve energy in storage systems include dynamic power management [4], power-aware cache management [6], power-aware prefetching strategy [8], software-directed power management scheme [9], and multi-speed settings strategy [13]. However, none of these techniques have addressed the energy conservation and performance issues of parallel storage systems with buffer disks.…”

A prefetching scheme for energy conservation in parallel disk systems

“…For example, optimizing compilers for multi-speed disks were developed to produce energy savings [17]. Our approach is quite different from these existing approaches because ours is focused on the architecture of parallel disk systems and the management of data on all the disks in a parallel disk system.…”