POSE: A mathematical and visual modelling tool to guide energy aware code optimisation

Future Generation Computer Systems

2019

This special issue of Future Generation Computer Systems contains four extended papers selected from the 7th International Workshop on Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems (PMBS 2016), held as part of the 28th International Conference for High Performance Computing, Networking, Storage and Analysis (SC 2016). These papers represent worldwide programmes of research committed to understanding application and architecture performance to enable post-peta-scale computational science.

Section: Challenges Of Exascalementioning

confidence: 99%

Performance Modeling, Benchmarking and Simulation of High Performance Computing Systems

Future Generation Computer Systems

2019

“…POSE has a number of potential uses that we intend to explore in the future. First, we plan to revisit the use of frequency scaling and P-state selection to identify whether POSE can highlight additional opportunities for power-optimisation [35]. Second, we wish to use POSE to investigate the use of accelerator architectures; we believe GPU and FPGA architectures may offer greater opportunities for power optimisation compared to the x86-64 architectures outlined in this article [14].…”

Section: Future Workmentioning

confidence: 99%

The Power-optimised Software Envelope

Roberts

ACM Trans. Archit. Code Optim.

Fahmy

et al. 2019

Advances in processor design have delivered performance improvements for decades. As physical limits are reached, refinements to the same basic technologies are beginning to yield diminishing returns. Unsustainable increases in energy consumption are forcing hardware manufacturers to prioritise energy efficiency in their designs. Research suggests that software modifications may be needed to exploit the resulting improvements in current and future hardware. New tools are required to capitalise on this new class of optimisation. In this article, we present the Power Optimised Software Envelope (POSE) model, which allows developers to assess the potential benefits of power optimisation for their applications. The POSE model is metric agnostic and in this article, we provide derivations using the established Energy-Delay Product metric and the novel Energy-Delay Sum and Energy-Delay Distance metrics that we believe are more appropriate for energy-aware optimisation efforts. We demonstrate POSE on three platforms by studying the optimisation characteristics of applications from the Mantevo benchmark suite. Our results show that the Pathfinder application has very little scope for power optimisation while TeaLeaf has the most, with all other applications in the benchmark suite falling between the two. Finally, we extend our POSE model with a formulation known as System Summary POSE-a meta-heuristic that allows developers to assess the scope a system has for energy-aware software optimisation independent of the code being run. CCS Concepts: • Computing methodologies → Modeling and simulation; Model development and analysis; Modeling methodologies; • Hardware → Power and energy; Power estimation and optimization;

“…Whereas the Et n parameter operates in a relative fashion, however, EDS and EDD parameters are based on absolute costs of consumption. The power drawn by active Taurus nodes ranges between 207.68 W and 345.33 W [18], meaning the magnitude of energy costs will be around 300 times greater than that of runtime. In order to compensate for this effect we scale the runtime cost by a factor of 300 before applying the 3:1 ratio, resulting in the parameters α = 1 and β = 3 × 300 = 900.…”

Section: Case Studymentioning

confidence: 99%

“…Choi et al proposed the Energy Roofline model to identify the algorithmic conditions needed for trade-offs between runtime and energy [5]. Similarly, in previous work we developed the Power Optimised Software Envelope model to assess the scope a code has for power optimisations on any given platform [18].…”

Section: Related Workmentioning

confidence: 99%

Metrics for Energy-Aware Software Optimisation

Roberts

Lecture Notes in Computer Science

Fahmy

et al. 2017

A note on versions:The version presented here may differ from the published version or, version of record, if you wish to cite this item you are advised to consult the publisher's version. Please see the 'permanent WRAP url' above for details on accessing the published version and note that access may require a subscription.For more information, please contact the WRAP Team at: wrap@warwick.ac.uk Metrics for Energy-Aware Software OptimisationStephen I. Roberts, Steven A. Wright, Suhaib A. Fahmy, and Stephen A. Jarvis University of Warwick, U.K. S.I.Roberts@warwick.ac.uk Abstract. Energy consumption is rapidly becoming a limiting factor in scientific computing. As a result, hardware manufacturers increasingly prioritise energy efficiency in their processor designs. Performance engineers are also beginning to explore software optimisation and hardware/software co-design as a means to reduce energy consumption. Energy efficiency metrics developed by the hardware community are often re-purposed to guide these software optimisation efforts. In this paper we argue that established metrics, and in particular those in the Energy Delay Product (Et n ) family, are unsuitable for energy-aware software optimisation. A good metric should provide meaningful values for a single experiment, allow fair comparison between experiments, and drive optimisation in a sensible direction. We show that Et n metrics are unable to fulfil these basic requirements and present suitable alternatives for guiding energy-aware software optimisation. We finish with a practical demonstration of the utility of our proposed metrics.