{"110041":{"#nid":"110041","#data":{"type":"event","title":"Special CSE Seminar: Prof. Robert van de Geijn","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003EProf. Robert van de Geijn\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003E\u0022Design by Transformation - Application to Dense Linear Algebra Libraries\u0022\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EAbstract\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003EThe FLAME project has yielded modern alternatives to LAPACK and related effort.\u0026nbsp; An attractive feature of this effort is the complete vertical integration of the entire software stack, starting with low level kernels that support the BLAS and finishing with a new distributed memory library, Elemental.\u0026nbsp; In between are layers that target a single core, multicore, and multiGPU architectures. What this now enables is a new approach where libraries are viewed not as instantiations in code but instead as a repository of algorithms, knowledge about those algorithm, and knowledge about target architectures. Representations in code are then mechanically generated by a tool that performs optimizations for a given architecture by applying high-level transformations much like a human expert would.\u0026nbsp; We discuss how this has been used to mechanically generate tens of thousands of different distributed memory implementations given a single sequential algorithm.\u0026nbsp; By attaching cost functions to the component operations, a highly optimized implementation is chosen by the tool that invariably matches or exceeds the performance of implementations by human experts. We call this approach Design by Transformation (DxT).\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EBio\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003ERobert van de Geijn is a Professor of Computer Science and member of the Institute for Computational Engineering and Sciences at UT-Austin. He received his Ph.D. in Applied Mathematics from the University of Maryland.\u0026nbsp; His interests are in linear algebra libraries, scientific computing, parallel computing, and formal derivation of programs. He has written more than a hundred refereed articles and several books.\u003C\/p\u003E\u003Cp\u003EThis work is in collaboration with Bryan Marker, Don Batory, Jack Poulson, and Andy Terrell.\u003C\/p\u003E","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Design by Transformation: Application to Dense Linear Algebra Libraries"}],"uid":"27439","created_gmt":"2012-02-17 09:27:59","changed_gmt":"2016-10-08 01:58:05","author":"Lometa Mitchell","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2012-02-21T10:00:00-05:00","event_time_end":"2012-02-21T11:00:00-05:00","event_time_end_last":"2012-02-21T11:00:00-05:00","gmt_time_start":"2012-02-21 15:00:00","gmt_time_end":"2012-02-21 16:00:00","gmt_time_end_last":"2012-02-21 16:00:00","rrule":null,"timezone":"America\/New_York"},"extras":["free_food"],"groups":[{"id":"47223","name":"College of Computing"},{"id":"50877","name":"School of Computational Science and Engineering"}],"categories":[],"keywords":[],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1795","name":"Seminar\/Lecture\/Colloquium"}],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003E\u003Ca href=\u0022mailto:richie@cc.gatech.edu\u0022\u003ERich Vuduc\u003C\/a\u003E\u003C\/p\u003E","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}