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

Owing to data-intensive large-scale applications, distributed computationsystems have gained significant recent interest, due to their ability ofrunning such tasks over a large number of commodity nodes in a time efficientmanner. One of the major bottlenecks that adversely impacts the time efficiencyis the computational heterogeneity of distributed nodes, often limiting thetask completion time due to the slowest worker.  In this paper, we first present a lower bound on the expected computationtime based on the work-conservation principle. We then present our approach ofwork exchange to combat the latency problem, in which faster workers can bereassigned additional leftover computations that were originally assigned toslower workers. We present two variations of the work exchange approach: a)when the computational heterogeneity knowledge is known a priori; and b) whenheterogeneity is unknown and is estimated in an online manner to assign tasksto distributed workers. As a baseline, we also present and analyze the use ofan optimized Maximum Distance Separable (MDS) coded distributed computationscheme over heterogeneous nodes. Simulation results also compare the proposedapproach of work exchange, the baseline MDS coded scheme and the lower boundobtained via work-conservation principle. We show that the work exchange schemeachieves time for computation which is very close to the lower bound withlimited coordination and communication overhead even when the knowledge aboutheterogeneity levels is not available.

Combating Computational Heterogeneity in Large-Scale Distributed Computing via Work Exchange