{"674461":{"#nid":"674461","#data":{"type":"event","title":"PhD Defense by Lechen Yu","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003ETitle\u003C\/strong\u003E: Dynamic Analysis of Data Inconsistencies and Data Races in OpenMP Programs\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EDate\u003C\/strong\u003E: Monday, May 6, 2024\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003ETime\u003C\/strong\u003E: 2:00 pm - 4:00 pm ET\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003ELocation\u003C\/strong\u003E: Klaus 3402 and Virtual (\u003Ca href=\u0022https:\/\/gatech.zoom.us\/j\/3092779704?pwd=SDVZaTlkTFpsZDRoYUU3Y0pzdmNvdz09\u0022\u003Ehttps:\/\/gatech.zoom.us\/j\/3092779704?pwd=SDVZaTlkTFpsZDRoYUU3Y0pzdmNvdz09\u003C\/a\u003E)\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003ELechen Yu\u003C\/strong\u003E\u003C\/p\u003E\r\n\r\n\u003Cp\u003EPh.D. Student\u003C\/p\u003E\r\n\r\n\u003Cp\u003ESchool of Computer Science\u003C\/p\u003E\r\n\r\n\u003Cp\u003EGeorgia Institute of Technology\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003ECommittee\u003C\/strong\u003E:\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Vivek Sarkar (Advisor) \u2013 School of Computer Science, Georgia Institute of Technology\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Santosh Pande \u2013 School of Computer Science, Georgia Institute of Technology\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Qirun Zhang \u2013 School of Computer Science, Georgia Institute of Technology\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Hyesoon Kim \u2013 School of Computer Science, Georgia Institute of Technology\u003C\/p\u003E\r\n\r\n\u003Cp\u003EDr. Tiago Cogumbreiro \u2013 Department of Computer Science, University of Massachusetts Boston\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EAbstract\u003C\/strong\u003E:\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003EOpenMP is a popular intra-node parallel programming model that supports several\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eproblem decomposition approaches, including task parallelism, data parallelism, and heterogeneous\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eparallelism. When OpenMP introduces new parallel paradigms or features, it must\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eensure that these additions align with the existing constructs to maintain consistency and\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eavoid unspecified behaviors. New features can result in revisions to the behavior of existing\u003C\/p\u003E\r\n\r\n\u003Cp\u003Econstructs, which may in turn require programmers to re-evaluate their understanding of\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eexisting constructs and adapt their code accordingly. Consequently, writing correct OpenMP\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eprograms can be challenging even for experienced programmers.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003ETo alleviate the intricacy of writing correct OpenMP programs, this proposal\u0026nbsp;outlines\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eseveral dynamic analysis techniques that help programmers identify programming errors in\u003C\/p\u003E\r\n\r\n\u003Cp\u003EOpenMP programs. First, we describe various studies on device offloading, a recent OpenMP\u003C\/p\u003E\r\n\r\n\u003Cp\u003Efeature still in its developmental phase. Our studies reveal discrepancies between the LLVM\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eimplementation and the intended runtime behavior of device offloading. Additionally,\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eerroneous usage of device offloading constructs can lead to an assortment of memory\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eanomalies. Since these memory anomalies can generate disparities between host variables\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eand their corresponding counterparts on accelerator devices, we classify such bugs as data\u003C\/p\u003E\r\n\r\n\u003Cp\u003Einconsistencies. By establishing permissible memory accesses on the host and accelerator,\u003C\/p\u003E\r\n\r\n\u003Cp\u003Ethis proposal\u0026nbsp;introduces a dynamic approach to detect data inconsistencies automatically.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe dynamic approach leverages a per-variable state transition model, which can be used to\u003C\/p\u003E\r\n\r\n\u003Cp\u003Eestablish the validity of the memory location before executing any memory accesses. Beyond\u003C\/p\u003E\r\n\r\n\u003Cp\u003Edata inconsistencies, this proposal\u0026nbsp;also delves into novel dynamic approaches for identifying\u003C\/p\u003E\r\n\r\n\u003Cp\u003Edata races in OpenMP programs. By extending the SPD3 race detection algorithm, originally\u003C\/p\u003E\r\n\r\n\u003Cp\u003Edesigned for async-finish task parallelism, our dynamic race detection approach can handle a\u003C\/p\u003E\r\n\r\n\u003Cp\u003Elarge subset of parallel constructs in OpenMP, thereby checking more precise happens-before\u003C\/p\u003E\r\n\r\n\u003Cp\u003Erelations among tasks relative to existing per-thread vector-clock-based approaches.\u003C\/p\u003E\r\n","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EDynamic Analysis of Data Inconsistencies and Data Races in OpenMP Programs\u003C\/p\u003E\r\n","format":"limited_html"}],"field_summary_sentence":[{"value":"Dynamic Analysis of Data Inconsistencies and Data Races in OpenMP Programs"}],"uid":"27707","created_gmt":"2024-04-30 22:15:52","changed_gmt":"2024-04-30 22:16:23","author":"Tatianna Richardson","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2024-05-06T14:00:00-04:00","event_time_end":"2024-05-06T16:00:00-04:00","event_time_end_last":"2024-05-06T16:00:00-04:00","gmt_time_start":"2024-05-06 18:00:00","gmt_time_end":"2024-05-06 20:00:00","gmt_time_end_last":"2024-05-06 20:00:00","rrule":null,"timezone":"America\/New_York"},"location":" Klaus 3402 and Virtual ","extras":[],"groups":[{"id":"221981","name":"Graduate Studies"}],"categories":[],"keywords":[{"id":"100811","name":"Phd Defense"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1788","name":"Other\/Miscellaneous"}],"invited_audience":[{"id":"78771","name":"Public"}],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}