Optimising Quantisation Noise in Energy Measurement

Langdon, William B.; Petke, Justyna; Bruce, Bobby R.

doi:10.1007/978-3-319-45823-6_23

Cited by 9 publications

(7 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…by removing bugs [8,9,10,11,12,13,14,15,16] or adding to its abilities [17,18,19,20,21,22]. Non-functional improvements that have been considered or results reported include: faster code [23,24], code which uses less energy [25,26,27,28,29,30,31,32,33,34] or less memory [35], and automatic parallelisation [36,37,38] and automatic porting [39] and embedded systems [40,41,25,42,43,44,45] as well as refactorisation [46], reverse engineering [47,48] and software product lines [49,50]. There is very much a GI flavour in the air with a three-fold increase in GI publications (as measured by GI papers in the genetic programming bibliography) since the first GI workshop [51] was first mooted (October, 7 2014) 1 .…”

Section: Genetic Improvementmentioning

confidence: 99%

Visualising the Search Landscape of the Triangle Program

Langdon¹,

Veerapen

Ochoa

2017

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Abstract. High order mutation analysis of a software engineering benchmark, including schema and local optima networks, suggests program improvements may not be as hard to find as is often assumed. 1) Bitwise genetic building blocks are not deceptive and can lead to all global optima. 2) There are many neutral networks, plateaux and local optima, nevertheless in most cases near the human written C source code there are hill climbing routes including neutral moves to solutions.

show abstract

Section: Genetic Improvementmentioning

confidence: 99%

Visualising the Search Landscape of the Triangle Program

Langdon¹,

Veerapen

Ochoa

2017

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

show abstract

“…In some previous GI work we had used actual run time, e.g. [40]. However in [40], we evolved subroutines which could be called directly by our GI system, whereas here we will test complete programs and so need the unix process time.…”

Section: Counting Instructions With Perf Stat -E Instructions -Xmentioning

confidence: 99%

“…[40]. However in [40], we evolved subroutines which could be called directly by our GI system, whereas here we will test complete programs and so need the unix process time. Also runtime is notoriously noisy and we had previously found success using the unix perf tool, e.g.…”

Section: Counting Instructions With Perf Stat -E Instructions -Xmentioning

confidence: 99%

Genetic Improvement of LLVM Intermediate Representation

Langdon¹,

Al-Subaihin²,

Blot³

et al. 2023

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

Evolving LLVM IR is widely applicable, with LLVM Clang offering support for an increasing range of computer hardware and programming languages. Local search mutations are used to hill climb industry C code released to support geographic open standards: Open Location Code (OLC) from Google and Uber's Hexagonal Hierarchical Spatial Index (H3), giving up to two percent speed up on compiler optimised code.

show abstract

“…Due to inevitable overheads associated with sending energy measurement start and stop commands over a network [36], [37], we chose applications that have a non-trivial execution time, which we have defined to be greater than 5 seconds. The larger the execution time, the smaller the overheads are as a percentage of total energy consumption.…”

Section: Selection Criteriamentioning

confidence: 99%

Approximate Oracles and Synergy in Software Energy Search Spaces

Bruce

Petke

Harman

et al. 2019

IIEEE Trans. Software Eng.

Self Cite

View full text Add to dashboard Cite

Abstract-Reducing the energy consumption of software systems though optimisations techniques such as genetic improvement is gaining interest. However, efficient and effective improvement of software systems requires a better understanding of the code-change search space. One important choice practitioners have is whether to preserve the system's original output or permit approximation with each scenario having its own search space characteristics. When output preservation is a hard constraint, we report that the maximum energy reduction achievable by the modification operators is 2.69% (0.76% on average). By contrast, this figure increases dramatically to 95.60% (33.90% on average) when approximation is permitted, indicating the critical importance of approximate output quality assessment for code optimisation. We investigate synergy, a phenomenon that occurs when simultaneously applied source code modifications produce an effect greater than their individual sum. Our results reveal that 12.0% of all joint code modifications produced such a synergistic effect though 38.5% produce an antagonistic interaction in which simultaneously applied modifications are less effective than when applied individually. This highlights the need for more advanced search-based approaches.

show abstract

Optimising Quantisation Noise in Energy Measurement

Cited by 9 publications

References 7 publications

Visualising the Search Landscape of the Triangle Program

Visualising the Search Landscape of the Triangle Program

Genetic Improvement of LLVM Intermediate Representation

Approximate Oracles and Synergy in Software Energy Search Spaces

Contact Info

Product

Resources

About