English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Significant efforts have been devoted to choosing the best configuration of acomputing system to run an application energy efficiently. However, availabletuning approaches mainly focus on homogeneous systems and are inextensible forheterogeneous systems which include several components (e.g., CPUs, GPUs) withdifferent architectures. This study proposes a holistic tuning approach calledREOH using probabilistic network to predict the most energy-efficientconfiguration (i.e., which platform and its setting) of a heterogeneous systemfor running a given application. Based on the computation and communicationpatterns from Berkeley dwarfs, we conduct experiments to devise the trainingset including 7074 data samples covering varying application patterns andcharacteristics. Validating the REOH approach on heterogeneous systemsincluding CPUs and GPUs shows that the energy consumption by the REOH approachis close to the optimal energy consumption by the Brute Force approach whilesaving 17% of sampling runs compared to the previous (homogeneous) approachusing probabilistic network. Based on the REOH approach, we develop anopen-source energy-optimizing runtime framework for selecting an energyefficient configuration of a heterogeneous system for a given application atruntime.

REOH: Using Probabilistic Network for Runtime Energy Optimization of Heterogeneous Systems