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

High intensive computation applications can usually take days to months tofinish an execution. During this time, it is common to have variations of theavailable resources when considering that such hardware is usually shared amonga plurality of researchers/departments within an organization. On the otherhand, High Performance Clusters can take advantage of Cloud Computing burstingtechniques for the execution of applications together with the on-premiseresources. In order to meet deadlines, high intensive computationalapplications can use the Cloud to boost their performance when they are dataand task parallel. This article presents an ongoing work towards the use ofextended resources of an HPC execution platform together with Cloud. We proposean unified view of such heterogeneous environments and a method that monitors,predicts the application execution time, and dynamically shifts part of thedomain -- previously running in local HPC hardware -- to be computed on theCloud, meeting then a specific deadline. The method is exemplified along with aseismic application that, at runtime, adapts itself to move part of theprocessing to the Cloud (in a movement called bursting) and also auto-scales(the moved part) over cloud nodes. Our preliminary results show that there isan expected overhead for performing this movement and for synchronizingresults, but our outcomes demonstrate it is an important feature for meetingdeadlines in the case an on-premise cluster is overloaded or cannot provide thecapacity needed for a particular project.

A Self-adaptive Auto-scaling Method for Scientific Applications on HPC Environments and Clouds