{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/101460"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/101460","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Optimizing interactive analytics engines for heterogeneous clusters","abstract":"This thesis targets the growing area of interactive data analytics engines. It builds upon a system called Getafix, an intelligent data replication and placement algorithm, and optimizes Getafix for running mixed queries over a heterogeneous cluster. The new algorithm is called Getafix-H, a cluster aware version of Getafix replication algorithm, with built-in optimizations for segment balancing and cluster auto-tiering. We integrated Getafix-H as an extension to Getafix inside Druid, a modern open-source interactive data analytics engine. We present experimental results using workloads from Yahoo!’s production Druid cluster. Compared to Getafix, Getafix-H improves the tail latency by 18% and reduces memory usage by up to 27% (2-3X improvement over Scarlett). In presence of stragglers, Getafix-H improves tail latency by 55% and reduces memory usage by upto 20% compared to Getafix. Getafix-H enables sysadmins to auto-tier a heterogeneous cluster with the tiering accuracy of up to 80%.","abstract_html":"This thesis targets the growing area of interactive data analytics engines. It builds upon a system called Getafix, an intelligent data replication and placement algorithm, and optimizes Getafix for running mixed queries over a heterogeneous cluster. The new algorithm is called Getafix-H, a cluster aware version of Getafix replication algorithm, with built-in optimizations for segment balancing and cluster auto-tiering. We integrated Getafix-H as an extension to Getafix inside Druid, a modern open-source interactive data analytics engine. We present experimental results using workloads from Yahoo!’s production Druid cluster. Compared to Getafix, Getafix-H improves the tail latency by 18% and reduces memory usage by up to 27% (2-3X improvement over Scarlett). In presence of stragglers, Getafix-H improves tail latency by 55% and reduces memory usage by upto 20% compared to Getafix. Getafix-H enables sysadmins to auto-tier a heterogeneous cluster with the tiering accuracy of up to 80%.","abstract_has_math":false,"creators":["Raina, Ashwini"],"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":2018,"date_issued":"2018-09-27T16:17:16Z","date_published":"2018-09-27T16:17:16Z","updated_at":"2026-07-22T22:24:40Z","subjects":["Real-time analytics, data replication"],"languages":["en"],"rights":["Copyright 2018 Ashwini Raina"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/101460","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":["Raina, Ashwini"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2018-09-27T16:17:16Z","2018-05-09","2018-08"]},{"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":["Real-time analytics, data replication"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2018 Ashwini Raina"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/101460"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["This thesis targets the growing area of interactive data analytics engines. It builds upon a system called Getafix, an intelligent data replication and placement algorithm, and optimizes Getafix for running mixed queries over a heterogeneous cluster. The new algorithm is called Getafix-H, a cluster aware version of Getafix replication algorithm, with built-in optimizations for segment balancing and cluster auto-tiering. We integrated Getafix-H as an extension to Getafix inside Druid, a modern open-source interactive data analytics engine. We present experimental results using workloads from Yahoo!’s production Druid cluster. Compared to Getafix, Getafix-H improves the tail latency by 18% and reduces memory usage by up to 27% (2-3X improvement over Scarlett). In presence of stragglers, Getafix-H improves tail latency by 55% and reduces memory usage by upto 20% compared to Getafix. Getafix-H enables sysadmins to auto-tier a heterogeneous cluster with the tiering accuracy of up to 80%.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2018-09-27 without embargo terms","The student, Ashwini Raina, accepted the attached license on 2018-05-08 at 12:00.","The student, Ashwini Raina, submitted this Thesis for approval on 2018-05-08 at 12:14.","This Thesis was approved for publication on 2018-05-09 at 12:46.","DSpace SAF Submission Ingestion Package generated from Vireo submission #12563 on 2018-09-27 at 10:43:43","Made available in DSpace on 2018-09-27T16:17:16Z (GMT). No. of bitstreams: 2 RAINA-THESIS-2018.pdf: 578295 bytes, checksum: 50e840637ea2e1e73dbcf2a17e240bb6 (MD5) LICENSE.txt: 4210 bytes, checksum: c8b606cf11f891caea015bfab8b34141 (MD5) Previous issue date: 2018-05-09"]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Optimizing interactive analytics engines for heterogeneous clusters"]}]}],"canonical_facts":{"dc:contributor":["Gupta, Indranil"],"dc:creator":["Raina, Ashwini"],"dc:date":["2018-09-27T16:17:16Z","2018-05-09","2018-08"],"dc:description":["This thesis targets the growing area of interactive data analytics engines. It builds upon a system called Getafix, an intelligent data replication and placement algorithm, and optimizes Getafix for running mixed queries over a heterogeneous cluster. The new algorithm is called Getafix-H, a cluster aware version of Getafix replication algorithm, with built-in optimizations for segment balancing and cluster auto-tiering. We integrated Getafix-H as an extension to Getafix inside Druid, a modern open-source interactive data analytics engine. We present experimental results using workloads from Yahoo!’s production Druid cluster. Compared to Getafix, Getafix-H improves the tail latency by 18% and reduces memory usage by up to 27% (2-3X improvement over Scarlett). In presence of stragglers, Getafix-H improves tail latency by 55% and reduces memory usage by upto 20% compared to Getafix. Getafix-H enables sysadmins to auto-tier a heterogeneous cluster with the tiering accuracy of up to 80%.","Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2018-09-27 without embargo terms","The student, Ashwini Raina, accepted the attached license on 2018-05-08 at 12:00.","The student, Ashwini Raina, submitted this Thesis for approval on 2018-05-08 at 12:14.","This Thesis was approved for publication on 2018-05-09 at 12:46.","DSpace SAF Submission Ingestion Package generated from Vireo submission #12563 on 2018-09-27 at 10:43:43","Made available in DSpace on 2018-09-27T16:17:16Z (GMT). No. of bitstreams: 2 RAINA-THESIS-2018.pdf: 578295 bytes, checksum: 50e840637ea2e1e73dbcf2a17e240bb6 (MD5) LICENSE.txt: 4210 bytes, checksum: c8b606cf11f891caea015bfab8b34141 (MD5) Previous issue date: 2018-05-09"],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/101460"],"dc:language":["en"],"dc:rights":["Copyright 2018 Ashwini Raina"],"dc:subject":["Real-time analytics, data replication"],"dc:title":["Optimizing interactive analytics engines for heterogeneous clusters"],"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":"2026-07-22T22:24:40Z"}