{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/90610"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/90610","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"New techniques to lower the tail latency in stream processing systems","abstract":"Over the past decade, the demand for real time processing of huge amount of streaming data has emerged and grown rapidly. Apache Storm, Apache Flink, Samza and many other stream processing frameworks have been proposed and implemented to meet this need. Although lots of effort has been made to reduce the average latency of stream processing systems, how to shorten their tail latency has received little attention. This thesis presents a series of novel techniques for reducing the tail latency in stream processing systems like Apache Storm. Concretely, we present three mechanisms: (1) adaptive timeout coupled with selective replay to catch straggler tuples; (2) shared queues among different tasks of the same operator to reduce overall queueing delay; (3) latency feedback-based load balancing, intended to mitigate heterogenous scenarios. We have implemented these techniques in Apache Storm, and present experimental results using sets of micro-benchmarks as well as two topologies from Yahoo! Inc. Our results show improvement in tail latency in the range of 2%-72.9%.","abstract_html":"Over the past decade, the demand for real time processing of huge amount of streaming data has emerged and grown rapidly. Apache Storm, Apache Flink, Samza and many other stream processing frameworks have been proposed and implemented to meet this need. Although lots of effort has been made to reduce the average latency of stream processing systems, how to shorten their tail latency has received little attention. This thesis presents a series of novel techniques for reducing the tail latency in stream processing systems like Apache Storm. Concretely, we present three mechanisms: (1) adaptive timeout coupled with selective replay to catch straggler tuples; (2) shared queues among different tasks of the same operator to reduce overall queueing delay; (3) latency feedback-based load balancing, intended to mitigate heterogenous scenarios. We have implemented these techniques in Apache Storm, and present experimental results using sets of micro-benchmarks as well as two topologies from Yahoo! Inc. Our results show improvement in tail latency in the range of 2%-72.9%.","abstract_has_math":false,"creators":["Du, Guangxiang"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Computer Science","degree_department":null,"school":null,"contributors":["Gupta, Indranil"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2016,"date_issued":"2016-07-07T19:54:34Z","date_published":"2016-07-07T19:54:34Z","updated_at":"2026-07-22T22:26:34Z","subjects":["tail latency","stream processing systems"],"languages":["en"],"rights":["Copyright 2016 Guangxiang Du"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/90610","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Gupta, Indranil"]},{"key":"dc:creator","label":"Author","values":["Du, Guangxiang"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2016-07-07T19:54:34Z","2016-04-22","2016-05"]},{"key":"dc:type","label":"Dc Type","values":["text"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Computer Science"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["tail latency","stream processing systems"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2016 Guangxiang Du"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/90610"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Over the past decade, the demand for real time processing of huge amount of streaming data has emerged and grown rapidly. Apache Storm, Apache Flink, Samza and many other stream processing frameworks have been proposed and implemented to meet this need. Although lots of effort has been made to reduce the average latency of stream processing systems, how to shorten their tail latency has received little attention. This thesis presents a series of novel techniques for reducing the tail latency in stream processing systems like Apache Storm. Concretely, we present three mechanisms: (1) adaptive timeout coupled with selective replay to catch straggler tuples; (2) shared queues among different tasks of the same operator to reduce overall queueing delay; (3) latency feedback-based load balancing, intended to mitigate heterogenous scenarios. We have implemented these techniques in Apache Storm, and present experimental results using sets of micro-benchmarks as well as two topologies from Yahoo! Inc. Our results show improvement in tail latency in the range of 2%-72.9%.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2016-07-07 without embargo terms","The student, Guangxiang Du, accepted the attached license on 2016-04-21 at 12:00.","The student, Guangxiang Du, submitted this Thesis for approval on 2016-04-21 at 12:05.","This Thesis was approved for publication on 2016-04-22 at 10:07.","DSpace SAF Submission Ingestion Package generated from Vireo submission #9404 on 2016-07-07 at 13:32:16","Made available in DSpace on 2016-07-07T19:54:34Z (GMT). No. of bitstreams: 2 DU-THESIS-2016.pdf: 887752 bytes, checksum: f5bfbf6b1ce1bc2b439cad4b9653c581 (MD5) LICENSE.txt: 4210 bytes, checksum: 3f5d8e6129c5e208f74bc8a562613078 (MD5) Previous issue date: 2016-04-22"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["New techniques to lower the tail latency in stream processing systems"]}]}],"canonical_facts":{"dc:contributor":["Gupta, Indranil"],"dc:creator":["Du, Guangxiang"],"dc:date":["2016-07-07T19:54:34Z","2016-04-22","2016-05"],"dc:description":["Over the past decade, the demand for real time processing of huge amount of streaming data has emerged and grown rapidly. Apache Storm, Apache Flink, Samza and many other stream processing frameworks have been proposed and implemented to meet this need. Although lots of effort has been made to reduce the average latency of stream processing systems, how to shorten their tail latency has received little attention. This thesis presents a series of novel techniques for reducing the tail latency in stream processing systems like Apache Storm. Concretely, we present three mechanisms: (1) adaptive timeout coupled with selective replay to catch straggler tuples; (2) shared queues among different tasks of the same operator to reduce overall queueing delay; (3) latency feedback-based load balancing, intended to mitigate heterogenous scenarios. We have implemented these techniques in Apache Storm, and present experimental results using sets of micro-benchmarks as well as two topologies from Yahoo! Inc. Our results show improvement in tail latency in the range of 2%-72.9%.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2016-07-07 without embargo terms","The student, Guangxiang Du, accepted the attached license on 2016-04-21 at 12:00.","The student, Guangxiang Du, submitted this Thesis for approval on 2016-04-21 at 12:05.","This Thesis was approved for publication on 2016-04-22 at 10:07.","DSpace SAF Submission Ingestion Package generated from Vireo submission #9404 on 2016-07-07 at 13:32:16","Made available in DSpace on 2016-07-07T19:54:34Z (GMT). No. of bitstreams: 2 DU-THESIS-2016.pdf: 887752 bytes, checksum: f5bfbf6b1ce1bc2b439cad4b9653c581 (MD5) LICENSE.txt: 4210 bytes, checksum: 3f5d8e6129c5e208f74bc8a562613078 (MD5) Previous issue date: 2016-04-22"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/90610"],"dc:language":["en"],"dc:rights":["Copyright 2016 Guangxiang Du"],"dc:subject":["tail latency","stream processing systems"],"dc:title":["New techniques to lower the tail latency in stream processing systems"],"dc:type":["text"],"thesis:degree_discipline":["Computer Science"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:26:34Z"}