{"id":185,"date":"2020-03-20T17:55:19","date_gmt":"2020-03-20T15:55:19","guid":{"rendered":"https:\/\/www.inf.telecom-sudparis.eu\/acmes\/jobs_cpt\/replaying-execution-trace-with-a-performance-model\/"},"modified":"2020-04-14T10:42:58","modified_gmt":"2020-04-14T08:42:58","slug":"replaying-execution-trace-with-a-performance-model","status":"publish","type":"job_cpt","link":"https:\/\/www.inf.telecom-sudparis.eu\/acmes\/jobs_cpt\/replaying-execution-trace-with-a-performance-model\/","title":{"rendered":"Replaying execution trace with a performance model"},"content":{"rendered":"<p>The Parallel &amp; Distributed systems group is looking for an internship.<\/p>\n<h3>Context<\/h3>\n<p>Parallel applications can be analyzed using tracing tools such as <a href=\"http:\/\/eztrace.gforge.inria.fr\/\">EZTrace<\/a>. These tools collect events (eg. calls to MPI functions, or OpenMP constructs, etc.) during the application execution, and the resulting execution traces can be analyzed to reveal performance bugs.<\/p>\n<p>As part of a collaborative research project, we develop <i>EasyTraceAnalyzer<\/i>, a generic trace analysis tool that processes various kinds of traces, and implement several performance analysis techniques (for instance, bottleneck detection).<\/p>\n<h3>Goal of this internship<\/h3>\n<p>This internship aims at designing a new performance analysis method that would predict the behavior of a parallel application if the performance of one of the platform components (eg. network, parallel filesystem, etc.) were different.<\/p>\n<p>The main tasks of the internship are the following:<\/p>\n<ul>\n<li><b>Extracting a performance model<\/b> from an execution trace by analyzing the execution trace (eg. MPI events, MPI-IO events, posix IO events, etc.)<\/li>\n<li><b>Applying a new performance model to a trace<\/b>. This will require to modify the execution trace to shorten\/expand the events related to the modified component, but also to spread to trace modification to the other threads\/processes that communicate together.<\/li>\n<li><b>Evaluating the performance prediction<\/b> on real applications<\/li>\n<\/ul>\n<p><b>Keywords:<\/b> HPC, MPI, performance analysis, performance modeling, performance prediction<\/p>\n<h3>Work conditions<\/h3>\n<ul>\n<li>Open-source development in C++<\/li>\n<li>The internship will take place at T\u00e9l\u00e9com SudParis at Palaiseau (in the same building as T\u00e9l\u00e9com Paris) &#8211; <a href=\"https:\/\/goo.gl\/maps\/o2g3wM2j39S2N4F36\">19 place Marguerite Perey, 91120 Palaiseau<\/a><\/li>\n<li>Due to the current confinement, the internship may start remotely<\/li>\n<\/ul>\n<h3>Contact<\/h3>\n<p><a href=\"https:\/\/trahay.wp.imtbs-tsp.eu\/\">Fran\u00e7ois Trahay<\/a> &lt;<a href=\"mailto:francois.trahay@telecom-sudparis.eu\">francois.trahay@telecom-sudparis.eu<\/a>&gt;, Associate professor<br \/>\n<a href=\"https:\/\/www.inf.telecom-sudparis.eu\/pds\/\"> Parallel &amp; Distributed Systems group<\/a><br \/>\n<a href=\"https:\/\/www.telecom-sudparis.eu\/\">T\u00e9l\u00e9com SudParis<\/a>, <a href=\"http:\/\/ip-paris.fr\/\">Institut Polytechnique de Paris<\/a>, <a href=\"http:\/\/samovar.telecom-sudparis.eu\/\">Samovar lab<\/a><\/p>\n","protected":false},"template":"","class_list":["post-185","job_cpt","type-job_cpt","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.inf.telecom-sudparis.eu\/acmes\/wp-json\/wp\/v2\/job_cpt\/185","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.inf.telecom-sudparis.eu\/acmes\/wp-json\/wp\/v2\/job_cpt"}],"about":[{"href":"https:\/\/www.inf.telecom-sudparis.eu\/acmes\/wp-json\/wp\/v2\/types\/job_cpt"}],"wp:attachment":[{"href":"https:\/\/www.inf.telecom-sudparis.eu\/acmes\/wp-json\/wp\/v2\/media?parent=185"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}