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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Emotion-aware computing can recognize, interpret, process, and simulate human affects. These programs in this area are compute-intensive applications, so they need to be executed in parallel. Loops usually have regular structures and programs spend significant amounts of time executing them, and thus loops are ideal candidates for exploiting the parallelism of sequential programs. However, it is difficult to decide which set of loops should be parallelized to improve program performance. The existing research is one-size-fits-all strategy and cannot guarantee to select profitable loops to be parallelized. This paper proposes a novel loop selection approach based on machine learning (ML-based) for selecting the profitable loops and paralleling them on multi-core by speculative multithreading (SpMT). It includes establishing sufficient training examples, building and applying prediction model to select profitable loops for speculative parallelization. Using the ML-based loop selection approach, an unseen emotion-aware sequential program can obtain a stable, much higher speedup than the one-size-fits-all approach. On Prophet, which is a generic SpMT processor to evaluate the performance of multithreaded programs, the novel loop selection approach is evaluated and reaches an average speedup of 1.87 on a 4-core processor. Experiment results show that the ML-based approach can obtain a significant increase in speedup, and Olden benchmarks deliver a better performance improvement of 6.70% on a 4-core than the one-size-fits-all approach.

Toward Emotion-Aware Computing: A Loop Selection Approach Based on Machine Learning for Speculative Multithreading