Del 14591 - Lecture Notes in Computer Science

Job Scheduling Strategies for Parallel Processing

27th International Workshop, JSSPP 2024, San Francisco, CA, USA, May 31, 2024, Revised Selected Papers

Häftad, Engelska, 2024

Av Dalibor Klusáček, Julita Corbalán, Gonzalo P. Rodrigo, Dalibor Klusacek, Julita Corbalan, Dalibor Klusáček, Dalibor Klusácek

819 kr

Beställningsvara. Skickas inom 10-15 vardagar

Fri frakt för medlemmar vid köp för minst 249 kr.

This book constitutes the refereed proceedings of the 27th International Workshop on Job Scheduling Strategies for Parallel Processing, JSSPP 2024, held in San Francisco, CA, USA, on May 31, 2024.The 10 full papers included in this book were carefully reviewed and selected from 15 submissions. The JSSPP 2024 covers several interesting problems within the resource management and scheduling domains.

Produktinformation

Utgivningsdatum2024-12-21
Mått155 x 235 x undefined mm
FormatHäftad
SpråkEngelska
SerieLecture Notes in Computer Science
Antal sidor197
FörlagSpringer International Publishing AG
ISBN9783031744297

Tillhör följande kategorier

Systemvetenskap och AI inom Data och it
Programvaruutveckling inom Data och it
Hårdvara inom Data och it
Artificiell intelligens inom Data och it
Referensverk och tvärvetenskap inom Samhalle och politik

.- Technical papers..- Real-life HPC Workload Trace Featuring Refined Job Runtime Estimates..- An Empirical Study of Machine Learning-based Synthetic Job Trace Generation Methods..- Clustering Based Job Runtime Prediction for Backfilling Using Classification..- Launchpad: Learning to Schedule Using Offline and Online RL Methods..- Radical-Cylon: A Heterogeneous Data Pipeline for Scientific Computing..- Evaluation of Heuristic Task-to-Thread Mapping Using Static and Dynamic Approaches..- Challenges in parallel matrix chain multiplication..- A node selection method for on-demand job execution with considering deadline constraints..- Maximizing Energy Budget Utilization Using Dynamic Power Cap Control..- Run your HPC jobs in Eco-Mode: revealing the potential of user-assisted power capping in supercomputing systems.