{"id":160425,"date":"2011-04-01T00:00:00","date_gmt":"2011-04-01T00:00:00","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/msr-research-item\/accurate-latency-estimation-in-a-distributed-event-processing-system-2\/"},"modified":"2018-10-16T20:17:36","modified_gmt":"2018-10-17T03:17:36","slug":"accurate-latency-estimation-in-a-distributed-event-processing-system-2","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/accurate-latency-estimation-in-a-distributed-event-processing-system-2\/","title":{"rendered":"Accurate Latency Estimation in a Distributed Event Processing System"},"content":{"rendered":"<div class=\"asset-content\">\n<p>A distributed event processing system consists of one or more nodes (machines), and can execute a directed acyclic graph (DAG) of operators called a dataflow (or query), over long-running high-event-rate data sources. An important component of such a system is cost estimation, which predicts or estimates the &#8220;goodness&#8221; of a given input, i.e., operator graph and\/or assignment of individual operators to nodes. Cost estimation is the foundation for solving many problems: optimization (plan selection and distributed operator placement), provisioning, admission control, and user reporting of system misbehavior.<\/p>\n<p>Latency is a significant user metric in many commercial real-time applications. Users are usually interested in quantiles of latency, such as worst-case or 99th percentile. However, existing cost estimation techniques for event-based dataflows use metrics that, while they may have the side-effect of being correlated with latency, do not directly or provably estimate latency. In this paper, we propose a new cost estimation technique using a metric called Mace (Maximum cumulative excess). Mace is provably equivalent to maximum system latency in a (potentially complex, multi-node) distributed event-based system. The close relationship to latency makes Mace ideal for addressing the problems described earlier. Experiments with real-world datasets on Microsoft StreamInsight deployed over 1-13 nodes in a data center validate our ability to closely estimate latency (within 4%), and the use of Mace for plan selection and distributed operator placement.<\/p>\n<\/div>\n<p><!-- .asset-content --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>A distributed event processing system consists of one or more nodes (machines), and can execute a directed acyclic graph (DAG) of operators called a dataflow (or query), over long-running high-event-rate data sources. An important component of such a system is cost estimation, which predicts or estimates the &#8220;goodness&#8221; of a given input, i.e., operator graph [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13547],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-160425","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-systems-and-networking","msr-locale-en_us"],"msr_publishername":"IEEE","msr_edition":"27th International Conference on Data Engineering (ICDE '11)","msr_affiliation":"","msr_published_date":"2011-04-01","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"27th International Conference on Data Engineering (ICDE '11)","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"206608","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","title":"latency-icde11.pdf","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/latency-icde11.pdf","id":206608,"label_id":0}],"msr_related_uploader":"","msr_attachments":[{"id":206608,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/latency-icde11.pdf"}],"msr-author-ordering":[{"type":"user_nicename","value":"badrishc","user_id":31166,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=badrishc"},{"type":"user_nicename","value":"jongold","user_id":32389,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=jongold"},{"type":"text","value":"Roger Barga","user_id":0,"rest_url":false},{"type":"text","value":"Mirek Riedewald","user_id":0,"rest_url":false},{"type":"user_nicename","value":"ivosan","user_id":32130,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=ivosan"},{"type":"user_nicename","value":"barga","user_id":0,"rest_url":false}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[957177],"msr_project":[170875],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":170875,"post_title":"Streams","post_name":"streams","post_type":"msr-project","post_date":"2011-11-21 13:31:30","post_modified":"2017-06-19 10:26:41","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/streams\/","post_excerpt":"In the streams research project, we propose novel architectures, efficient processing techniques, models, and applications to support time-oriented queries over real-time and offline data streams. Our current focus in the project centers around Trill, a high-performance streaming analytics engine that is now used across Microsoft. Our currect focus areas include efficient query processing, scale-out, resiliency, streaming state management, and unstructured data support.","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/170875"}]}}]},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/160425"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":1,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/160425\/revisions"}],"predecessor-version":[{"id":526019,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/160425\/revisions\/526019"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=160425"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=160425"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=160425"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=160425"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=160425"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=160425"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=160425"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=160425"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=160425"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=160425"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=160425"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=160425"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=160425"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=160425"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=160425"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=160425"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}