Inter-program Compilation for Disk Energy Reduction

Hom, Jerry; Kremer, Ulrich

doi:10.1007/978-3-540-28641-7_2

Cited by 3 publications

(5 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The interfering disk accesses resembles the original problem of unclustered disk requests in uniprogramming. A new scheduling technique, inverse barrier, was introduced to cluster disk accesses across applications [17]. The technique is a variation on barrier scheduling for parallel processing [23] and combines ideas from implicit co-scheduling for distributed systems [6] with the slotted ALOHA protocol [2,25].…”

Section: Related Workmentioning

confidence: 99%

Execution context optimization for disk energy

Hom

Kremer

2008

Proceedings of the 2008 International Conference on Compilers, Architectures and Synthesis for Embedded Systems

Self Cite

View full text Add to dashboard Cite

Power, energy, and thermal concerns have constrained embedded systems designs. Computing capability and storage density have increased dramatically, enabling the emergence of handheld devices from special to general purpose computing. In many mobile systems, the disk is among the top energy consumers. Many previous optimizations for disk energy have assumed uniprogramming environments. However, many optimizations degrade in multiprogramming because programs are unaware of other programs (execution context). We introduce a framework to make programs aware of and adapt to their runtime execution context.We evaluated real workloads by collecting user activity traces and characterizing the execution contexts. The study confirms that many users run a limited number of programs concurrently. We applied execution context optimizations to eight programs and tested ten combinations. The programs ran concurrently while the disk's power was measured. Our measurement infrastructure allows interactive sessions to be scripted, recorded, and replayed to compare the optimizations' effects against the baseline. Our experiments covered two write cache policies. For write-through, energy savings was in the range 3-63% with an average of 21%. For writeback, energy savings was in the range -33-61% with an average of 8%. In all cases, our optimizations incurred less than 1% performance penalty.

show abstract

Section: Related Workmentioning

confidence: 99%

Execution context optimization for disk energy

Hom

Kremer

2008

Proceedings of the 2008 International Conference on Compilers, Architectures and Synthesis for Embedded Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…A scheduling technique, inverse barrier, was proposed to synchronize disk accesses across active applications [5]. This mechanism is similar to implicit coscheduling for distributed systems [6].…”

Section: Related Workmentioning

confidence: 99%

“…Including execution context knowledge allows the applications to truly use the available resources rather than stick to a conservative assumption. In our framework, all interesting execution contexts are known at compile time and modeled as states of a finite state machine [5]. At runtime, process communication is necessary to inform active programs about changes in their execution context.…”

Section: Compiler / Runtime System Frameworkmentioning

confidence: 99%

Inter-program optimisations for disk energy reduction

Hom

Kremer²

2007

IJES

Self Cite

View full text Add to dashboard Cite

Previous work has shown that intra-program optimizations, i.e., optimizations performed on individual programs in isolation, can be very effective in reducing disk energy in streaming applications. This paper investigates the potential additional benefits of inter-program optimizations where sets of programs are optimized together. Experimental results on different subsets of three streaming applications show that 7-49% additional energy savings (27.3% on average) can be obtained with negligible performance penalties using two novel inter-program optimizations, namely execution context sensitive buffer size selection and inverse barrier synchronization. These figures were obtained via physical measurements on two laptop disks.

show abstract

“…A scheduling technique, inverse barrier, was proposed to synchronize disk accesses across active applications [7]. When a program accesses the disk, all other programs within its group are notified that the disk has been used, and they may decide to also use the disk.…”

Section: Related Workmentioning

confidence: 99%

“…Compiling with execution context knowledge allows the applications to truly use the available resources rather than stick to conservative assumptions. In our framework, all execution contexts are determined at compile time and modeled as states of a finite state machine [7]. At runtime, processes may passively or actively communicate about changes, such as programs starting or ending execution, in their execution context.…”

Section: Compiler / Runtime System Frameworkmentioning

confidence: 99%