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

Modern scientific applications predominantly run on large-scale computingplatforms, necessitating collaboration between scientific domain experts andhigh-performance computing (HPC) experts. While domain experts are oftenskilled in customizing domain-specific scientific computing routines, whichoften involves various matrix computations, HPC experts are essential forachieving efficient execution of these computations on large-scale platforms.This process often involves utilizing complex parallel computing librariestailored to specific matrix computation scenarios. However, the intricateprogramming procedure and the need for deep understanding in both applicationdomains and HPC poses significant challenges to the widespread adoption ofscientific computing. In this research, we observe that matrix computations canbe transformed into equivalent graph representations, and that by utilizinggraph processing engines, HPC experts can be freed from the burden ofimplementing efficient scientific computations. Based on this observation, weintroduce a graph engine-based scientific computing (Graph for Science)paradigm, which provides a unified graph programming interface, enabling domainexperts to promptly implement various types of matrix computations. Theproposed paradigm leverages the underlying graph processing engine to achieveefficient execution, eliminating the needs for HPC expertise in programminglarge-scale scientific applications. Our results show that the graphengine-based scientific computing paradigm achieves performance comparable tothe best-performing implementations based on existing parallel computinglibraries and bespoke implementations. Importantly, the paradigm greatlysimplifies the development of scientific computations on large-scale platforms,reducing the programming difficulty for scientists and facilitating broaderadoption of scientific computing.

Graph for Science: From API based Programming to Graph Engine based Programming for HPC