{"149421":{"#nid":"149421","#data":{"type":"event","title":"Ph.D. Thesis Proposal: Hrishikesh Amur","body":[{"value":"\u003Cp\u003ETitle: \u003Cstrong\u003EMemory-Efficient Distributed Parallel Frameworks using Compressed Buffer Trees\u003C\/strong\u003E\u003Cbr \/\u003E\u003Cbr \/\u003EHrishikesh Amur\u003Cbr \/\u003ESchool of Computer Science\u003Cbr \/\u003ECollege of Computing\u003Cbr \/\u003EGeorgia Institute of Technology\u003Cbr \/\u003E\u003Cbr \/\u003EDate: Friday August 31st, 2012\u003Cbr \/\u003E\u003Cstrong\u003ETime: 3:00PM - 5:00PM (EST) - UPDATED\u003C\/strong\u003E\u003Cbr \/\u003ELocation: KACB 3402\u003Cbr \/\u003E\u003Cbr \/\u003E\u003Cstrong\u003ECommittee:\u003C\/strong\u003E\u003C\/p\u003E\u003Cul\u003E\u003Cli\u003EDr. Karsten Schwan (Advisor, School of Computer Science, Georgia Tech)\u003C\/li\u003E\u003C\/ul\u003E\u003Cul\u003E\u003Cli\u003EDr. David Andersen (School of Computer Science, Carnegie Mellon University)\u003C\/li\u003E\u003C\/ul\u003E\u003Cul\u003E\u003Cli\u003EDr. Greg Ganger (School of Computer Science, Carnegie Mellon University)\u003C\/li\u003E\u003C\/ul\u003E\u003Cul\u003E\u003Cli\u003EDr. Ada Gavrilovska (School of Computer Science, Georgia Tech)\u003C\/li\u003E\u003C\/ul\u003E\u003Cul\u003E\u003Cli\u003EDr. Matthew Wolf (School of Computer Science, Georgia Tech)\u003C\/li\u003E\u003C\/ul\u003E\u003Cp\u003E\u003Cstrong\u003E\u003Cbr \/\u003E\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003Cbr \/\u003EMemory is a valuable commodity in datacenters. DRAM is expensive and an expensive consumer of power. With the number of cores per socket growing faster than the memory capacity per socket, memory is increasingly scarce. Given the rise of data-intensive computing, this focus on memory gains increased relevance. Data-intensive computing systems are primarily to designed to operate on large amounts of data from storage. However, in order to overcome the high latencies associated with disk access, applications commonly use memory for performance-sensitive data. Therefore, scarcity of memory can impact the performance of distributed applications signi\ufb01cantly.\u003Cbr \/\u003E\u003Cbr \/\u003EIn this thesis we introduce techniques for memory-ef\ufb01ciency without compromising performance. We introduce a novel data structure called the Compressed Buffer Tree (CBT) which stores data in memory-ef\ufb01cient form and allows computation to be executed on the data with high throughput. The CBT achieves memory-ef\ufb01ciency through the ef\ufb01cient application of data compression and of\ufb02oading of state of disk. We demonstrate the utility of the CBT through implementations of high-performance, memory-ef\ufb01cient runtimes for the following programming models, listed in order of increasing complexity:\u003C\/p\u003E\u003Cul\u003E\u003Cli\u003EMapReduce aggregation\u003C\/li\u003E\u003C\/ul\u003E\u003Cul\u003E\u003Cli\u003EGraph processing: MapReduce cannot handle dependencies in data or support iterative execution naturally; such dependencies are naturally captured by graphs. In distributed graph-processing libraries, communication can be handled either synchronously or asynchronously and can be message-passing-based or use shared memory. We show that the CBT can be used to implement runtimes for:\u003C\/li\u003E\u003C\/ul\u003E\u003Cp\u003E\u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp;\u0026nbsp;\u2013 a synchronous, message-passing model (Pregel)\u003C\/p\u003E\u003Cp\u003E\u0026nbsp;\u0026nbsp;\u0026nbsp; \u0026nbsp;\u0026nbsp;\u0026nbsp; \u2013 an asynchronous, shared-memory model (GraphLab)\u003C\/p\u003E","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Memory-Efficient Distributed Parallel Frameworks using Compressed Buffer Trees"}],"uid":"1","created_gmt":"2012-08-27 09:16:52","changed_gmt":"2016-10-08 01:59:37","author":"Jupiter","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2012-08-31T16:00:00-04:00","event_time_end":"2012-08-31T18:00:00-04:00","event_time_end_last":"2012-08-31T18:00:00-04:00","gmt_time_start":"2012-08-31 20:00:00","gmt_time_end":"2012-08-31 22:00:00","gmt_time_end_last":"2012-08-31 22:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"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\u003E\u003Ca href=\u0022mailto:amur@gatech.edu\u0022\u003EHrishikesh Amur\u003C\/a\u003E\u003C\/p\u003E","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}