{"601756":{"#nid":"601756","#data":{"type":"news","title":"Stanford CS Chair Talks Programming Models with Packed House","body":[{"value":"\u003Cp\u003EProfessor \u003Ca href=\u0022https:\/\/theory.stanford.edu\/~aiken\u0022\u003E\u003Cstrong\u003EAlex Aiken\u003C\/strong\u003E\u003C\/a\u003E delivered a School of Computer Science Distinguished Lecture on Wednesday, Jan. 31. His talk on programming models for large-scale parallel machines, \u003Cem\u003ELegion \u0026mdash; Programming Heterogeneous, Distributed Parallel Machine\u003C\/em\u003E, packed the auditorium in the Klaus Advanced Computing Building.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EAiken is the chair of computer science at Stanford and an ACM fellow. His research focuses on programming languages. The talk highlighted a new parallel programming system, known as \u003Cem\u003ELegion\u003C\/em\u003E, which works on heterogeneous, hierarchical, and distributed parallel machines.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cem\u003ELegion\u003C\/em\u003E is relevant in this world of supercomputers that require a lot of heterogeneity to function. A supercomputer can include central processing units (CPU), graphics processing units (GPU), field programming gate arrays, and more \u0026mdash; all with different memory sizes and speeds. Yet, because latency \u0026ndash; the delay between instructions for data transfer and the transfer itself \u0026ndash;is fixed in these systems, the power of computation is often faster than the ability to move data around. This can make it difficult to effectively program these high-end machines.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe three primary goals in programming the current generation of supercomputers are:\u003C\/p\u003E\r\n\r\n\u003Cul\u003E\r\n\t\u003Cli\u003EHigh performance\u003C\/li\u003E\r\n\t\u003Cli\u003EPerformance portability\u0026nbsp;\u003C\/li\u003E\r\n\t\u003Cli\u003EProgrammability\u003C\/li\u003E\r\n\u003C\/ul\u003E\r\n\r\n\u003Cp\u003EAlthough these goals can be met today, they come at a great cost. It can take a decade of software investment to develop one application, yet the machines change every three to four years. Currently, it is the programmer\u0026rsquo;s responsibility to keep up with what can often be tedious and time-consuming work.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;The real problem is not describing the parallelism; the problem is really the data and the movement of the data within these machines,\u0026rdquo; Aiken said.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe goal of \u003Cem\u003ELegion\u003C\/em\u003E is to shift some of this responsibility on to the programming system. When the organization of a program\u0026rsquo;s data is easily matched to a machine\u0026rsquo;s memory, it helps meet these performance goals. \u0026nbsp;\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cem\u003ELegion\u003C\/em\u003E is a programming model and runtime system that describes hierarchical organizations for data and computation at an abstract level. A separate mapping interface lets programmers control how data and computation are placed onto the memories and processors of a machine.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe results are effective. According to Aiken, \u0026nbsp;\u003Cem\u003ELegion\u003C\/em\u003E is competitive on a single graph, runs 20 times faster on GPUs, and is much faster than CPU systems.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u0026ldquo;You get a win on every single network we\u0026rsquo;ve tried just by doing partitioning and networking in a different way than is currently represented on other systems,\u0026rdquo; Aiken said.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EThe new programming model is already being used in the fields of chemistry and physics to model chemical reactions.\u003C\/p\u003E\r\n\r\n\u003Cp\u003EUltimately, Aiken presents a new way of looking at programming. If a programmer is willing to put in the effort to organize data up front, it will approve efficiency overall. \u0026ldquo;It\u0026rsquo;s all about program productivity. How much effort are you willing to expend to get the desired level of performance?\u0026rdquo; Aiken said.\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Cstrong\u003EWatch the lecture \u003Ca href=\u0022https:\/\/smartech.gatech.edu\/handle\/1853\/59347\u0022\u003Ehere\u003C\/a\u003E.\u003C\/strong\u003E\u003C\/p\u003E\r\n","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Alex Aiken from Stanford delivered a lecture."}],"uid":"34541","created_gmt":"2018-02-02 15:22:51","changed_gmt":"2018-02-23 18:01:25","author":"Tess Malone","boilerplate_text":"","field_publication":"","field_article_url":"","dateline":{"date":"2018-02-02T00:00:00-05:00","iso_date":"2018-02-02T00:00:00-05:00","tz":"America\/New_York"},"extras":[],"hg_media":{"601158":{"id":"601158","type":"image","title":"Alex Aiken","body":null,"created":"1516653832","gmt_created":"2018-01-22 20:43:52","changed":"1516653832","gmt_changed":"2018-01-22 20:43:52","alt":"Alex Aiken","file":{"fid":"229113","name":"Screen Shot 2018-01-22 at 3.43.08 PM.png","image_path":"\/sites\/default\/files\/images\/Screen%20Shot%202018-01-22%20at%203.43.08%20PM.png","image_full_path":"http:\/\/hg.gatech.edu\/\/sites\/default\/files\/images\/Screen%20Shot%202018-01-22%20at%203.43.08%20PM.png","mime":"image\/png","size":3033813,"path_740":"http:\/\/hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/Screen%20Shot%202018-01-22%20at%203.43.08%20PM.png?itok=Nx21XuDd"}}},"media_ids":["601158"],"groups":[{"id":"47223","name":"College of Computing"},{"id":"50875","name":"School of Computer Science"}],"categories":[],"keywords":[],"core_research_areas":[],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003ETess Malone, Communications Officer\u003C\/p\u003E\r\n\r\n\u003Cp\u003E\u003Ca href=\u0022mailto:tess.malone@cc.gatech.edu\u0022\u003Etess.malone@cc.gatech.edu\u003C\/a\u003E\u003C\/p\u003E\r\n","format":"limited_html"}],"email":["tess.malone@cc.gatech.edu"],"slides":[],"orientation":[],"userdata":""}}}