3 years ago

Malleable scheduling for flows of jobs and applications to MapReduce

Joel Wolf, Andrey Balmin, Kirsten Hildrum, Viswanath Nagarajan

Abstract

This paper provides a unified family of algorithms with performance guarantees for malleable scheduling problems on flows. A flow represents a set of jobs with precedence constraints. Each job has a speedup function that governs the rate at which work is done on the job as a function of the number of processors allocated to it. In our setting, each speedup function is linear up to some job-specific processor maximum. A key aspect of malleable scheduling is that the number of processors allocated to any job is allowed to vary with time. The overall objective is to minimize either the total cost (minisum) or the maximum cost (minimax) of the flows. Our approach handles a very general class of cost functions, and in particular provides the first constant-factor approximation algorithms for total and maximum weighted completion time. Our motivation for this work was scheduling in MapReduce, and we also provide experimental evaluations that show good practical performance.

Publisher URL: https://link.springer.com/article/10.1007/s10951-018-0576-y

DOI: 10.1007/s10951-018-0576-y

You might also like
Discover & Discuss Important Research

Keeping up-to-date with research can feel impossible, with papers being published faster than you'll ever be able to read them. That's where Researcher comes in: we're simplifying discovery and making important discussions happen. With over 19,000 sources, including peer-reviewed journals, preprints, blogs, universities, podcasts and Live events across 10 research areas, you'll never miss what's important to you. It's like social media, but better. Oh, and we should mention - it's free.

  • Download from Google Play
  • Download from App Store
  • Download from AppInChina

Researcher displays publicly available abstracts and doesn’t host any full article content. If the content is open access, we will direct clicks from the abstracts to the publisher website and display the PDF copy on our platform. Clicks to view the full text will be directed to the publisher website, where only users with subscriptions or access through their institution are able to view the full article.