Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226405595> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4226405595 abstract "Datacenters execute large computational jobs, which are composed of smaller tasks. A job completes when all its tasks finish, so stragglers -- rare, yet extremely slow tasks -- are a major impediment to datacenter performance. Accurately predicting stragglers would enable proactive intervention, allowing datacenter operators to mitigate stragglers before they delay a job. While much prior work applies machine learning to predict computer system performance, these approaches rely on complete labels -- i.e., sufficient examples of all possible behaviors, including straggling and non-straggling -- or strong assumptions about the underlying latency distributions -- e.g., whether Gaussian or not. Within a running job, however, none of this information is available until stragglers have revealed themselves when they have already delayed the job. To predict stragglers accurately and early without labeled positive examples or assumptions on latency distributions, this paper presents NURD, a novel Negative-Unlabeled learning approach with Reweighting and Distribution-compensation that only trains on negative and unlabeled streaming data. The key idea is to train a predictor using finished tasks of non-stragglers to predict latency for unlabeled running tasks, and then reweight each unlabeled task's prediction based on a weighting function of its feature space. We evaluate NURD on two production traces from Google and Alibaba, and find that compared to the best baseline approach, NURD produces 2--11 percentage point increases in the F1 score in terms of prediction accuracy, and 2.0--8.8 percentage point improvements in job completion time." @default.
- W4226405595 created "2022-05-05" @default.
- W4226405595 creator A5061121471 @default.
- W4226405595 creator A5065782554 @default.
- W4226405595 creator A5075192470 @default.
- W4226405595 creator A5080682322 @default.
- W4226405595 creator A5080833704 @default.
- W4226405595 date "2022-03-15" @default.
- W4226405595 modified "2023-10-17" @default.
- W4226405595 title "NURD: Negative-Unlabeled Learning for Online Datacenter Straggler Prediction" @default.
- W4226405595 doi "https://doi.org/10.48550/arxiv.2203.08339" @default.
- W4226405595 hasPublicationYear "2022" @default.
- W4226405595 type Work @default.
- W4226405595 citedByCount "0" @default.
- W4226405595 crossrefType "posted-content" @default.
- W4226405595 hasAuthorship W4226405595A5061121471 @default.
- W4226405595 hasAuthorship W4226405595A5065782554 @default.
- W4226405595 hasAuthorship W4226405595A5075192470 @default.
- W4226405595 hasAuthorship W4226405595A5080682322 @default.
- W4226405595 hasAuthorship W4226405595A5080833704 @default.
- W4226405595 hasBestOaLocation W42264055951 @default.
- W4226405595 hasConcept C119857082 @default.
- W4226405595 hasConcept C126838900 @default.
- W4226405595 hasConcept C127413603 @default.
- W4226405595 hasConcept C154945302 @default.
- W4226405595 hasConcept C183115368 @default.
- W4226405595 hasConcept C201995342 @default.
- W4226405595 hasConcept C2524010 @default.
- W4226405595 hasConcept C2780451532 @default.
- W4226405595 hasConcept C28719098 @default.
- W4226405595 hasConcept C33923547 @default.
- W4226405595 hasConcept C41008148 @default.
- W4226405595 hasConcept C46686674 @default.
- W4226405595 hasConcept C71924100 @default.
- W4226405595 hasConcept C76155785 @default.
- W4226405595 hasConcept C82876162 @default.
- W4226405595 hasConceptScore W4226405595C119857082 @default.
- W4226405595 hasConceptScore W4226405595C126838900 @default.
- W4226405595 hasConceptScore W4226405595C127413603 @default.
- W4226405595 hasConceptScore W4226405595C154945302 @default.
- W4226405595 hasConceptScore W4226405595C183115368 @default.
- W4226405595 hasConceptScore W4226405595C201995342 @default.
- W4226405595 hasConceptScore W4226405595C2524010 @default.
- W4226405595 hasConceptScore W4226405595C2780451532 @default.
- W4226405595 hasConceptScore W4226405595C28719098 @default.
- W4226405595 hasConceptScore W4226405595C33923547 @default.
- W4226405595 hasConceptScore W4226405595C41008148 @default.
- W4226405595 hasConceptScore W4226405595C46686674 @default.
- W4226405595 hasConceptScore W4226405595C71924100 @default.
- W4226405595 hasConceptScore W4226405595C76155785 @default.
- W4226405595 hasConceptScore W4226405595C82876162 @default.
- W4226405595 hasLocation W42264055951 @default.
- W4226405595 hasOpenAccess W4226405595 @default.
- W4226405595 hasPrimaryLocation W42264055951 @default.
- W4226405595 hasRelatedWork W1603068983 @default.
- W4226405595 hasRelatedWork W1987859285 @default.
- W4226405595 hasRelatedWork W1996541855 @default.
- W4226405595 hasRelatedWork W2051854463 @default.
- W4226405595 hasRelatedWork W2100760408 @default.
- W4226405595 hasRelatedWork W2129857763 @default.
- W4226405595 hasRelatedWork W3151529617 @default.
- W4226405595 hasRelatedWork W3179325884 @default.
- W4226405595 hasRelatedWork W3195168932 @default.
- W4226405595 hasRelatedWork W4281754391 @default.
- W4226405595 isParatext "false" @default.
- W4226405595 isRetracted "false" @default.
- W4226405595 workType "article" @default.