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

With the increasing amount of data available to scientists in disciplines asdiverse as bioinformatics, physics, and remote sensing, scientific workflowsystems are becoming increasingly important for composing and executingscalable data analysis pipelines. When writing such workflows, users need tospecify the resources to be reserved for tasks so that sufficient resources areallocated on the target cluster infrastructure. Crucially, underestimating atask's memory requirements can result in task failures. Therefore, users oftenresort to overprovisioning, resulting in significant resource wastage anddecreased throughput.  In this paper, we propose a novel online method that uses monitoring timeseries data to predict task memory usage in order to reduce the memory wastageof scientific workflow tasks. Our method predicts a task's runtime, divides itinto k equally-sized segments, and learns the peak memory value for eachsegment depending on the total file input size. We evaluate the prototypeimplementation of our method using workflows from the publicly availablenf-core repository, showing an average memory wastage reduction of 29.48%compared to the best state-of-the-art approach

Predicting Dynamic Memory Requirements for Scientific Workflow Tasks