{"65700":{"#nid":"65700","#data":{"type":"event","title":"Ph.D. Defense of Dissertation: Gong Zhang","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003EData and Application Migration in Cloud based Data\nCenters: Architectures and techniques\u003C\/strong\u003E\u003C\/p\u003E\n\n\n\n\n\n\u003Cp\u003E\u003Cstrong\u003EGong Zhang\u003C\/strong\u003E\u003Cbr \/\u003ESchool of Computer Science\u003Cbr \/\u003EGeorgia Institute of Technology\u003C\/p\u003E\n\n\u003Cp\u003E\u003Cstrong\u003ECommittee:\u003C\/strong\u003E\u003C\/p\u003E\n\n\n\n\u003Cp\u003EProf. Ling Liu (Advisor, School of Computer Science,\nGeorgia Tech)\u003Cbr \/\u003EProf. Sham Navathe (School of Computer Science, Georgia Tech)\u003Cbr \/\u003EProf. Calton Pu (School of Computer Science, Georgia Tech)\u003Cbr \/\u003EProf. Ed Omiecinski,\n(School of Computer Science, Georgia Tech)\u003Cbr \/\u003EProf. Joao Eduardo Ferreira,\n(Department of Computer Science, University of Sao Paulo)\u003C\/p\u003E\n\n\u003Cp\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003C\/p\u003E\n\n\u003Cp\u003EComputing and communication have continued to expedite\nthe growth of digital data and the complexity of applications. Today, the cost\nof managing and scaling hardware and software systems ranges from two to ten\ntimes the acquisition cost of the hardware and software systems. Such cost\ncontinues to increase as data grows and applications become more complex.\u003C\/p\u003E\n\n\n\n\u003Cp\u003EWe see an increasing demand on technologies for\ntransferring management burden from humans to software. Data migration and\napplication migration are popular technologies that enable computing and system\nmanagement to be autonomic and self-managing. \u003C\/p\u003E\n\n\n\n\u003Cp\u003EIn this dissertation, we examine important issues in\ndesigning and developing scalable architectures and techniques for efficient\nand effective data migration and application migration. The first contribution\nwe have made is to investigate the opportunity of automated data migration\nacross multi-tier storage systems. The significant IO improvement in Solid\nState Disks (SSD) over traditional rotational hard disks (HDD) motivates the\nintegration of SSD into existing storage hierarchy for enhanced performance. We\ndeveloped adaptive look-ahead data migration approach to effectively integrate\nSSD into the multi-tiered storage architecture. When using the fast and\nexpensive SSD tier to store the high temperature data (hot data) while placing\nthe relatively low temperature data (low data) in the HDD tier, one of the\nimportant functionality is to manage the migration of data as their access\npatterns are changed from hot to cold and vice versa. We designed and\nimplemented an adaptive lookahead data migration model that can dynamically\nadapt the data migration schedule to achieve the optimal migration\neffectiveness by taking into account of application specific characteristics\nand I\/O profiles as well as workload deadlines.\u003C\/p\u003E\n\n\n\n\u003Cp\u003EThe second main contribution we have made in this\ndissertation research is to address the challenge of ensuring reliability and\nbalancing loads across a network of computing nodes, managed in a decentralized\nservice computing system. We have developed a distributed workload migration\nscheme with controlled replication. It utilizes a shortcut-based optimization\nto increase the resilience of the system against various node failures and\nnetwork partition failures. In addition, we devise a dynamic load balancing\ntechnique to scale the system in anticipation of unexpected workload changes.\nOur approach is highly scalable under changing service workloads with moving\nhotspots and highly reliable in the presence of massive node failures.\u003C\/p\u003E\u003Cp\u003EThe third contribution in this dissertation research is\nto study how to simplify the management overheads in migrating large scale\nenterprise applications\/system from local data center to the Cloud. More and\nmore enterprises are moving some workloads from their local data centers to\nCloud, such as EC2, to reduce the cost of ownership and leverage the benefits\nprovided by Cloud based data centers. However, such migration process turns out\nto be non-trivial. By in-depth analysis of some popular multi-tier middleware\nsystems such as Hadoop, MySQL etc, we show the complexities and difficulties of\napplication migration process and propose an autonomic management framework for\nmigration configuration and migration validation, aiming at reducing typical\noperator errors, eliminating the risks of hidden migration pitfalls, and\nincreasing migration assurance. In this defense talk, I will give an overview\nof my dissertation research and then highlight in detail our most recent work\non application migration validation. \u003C\/p\u003E","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Data and Application Migration in Cloud based Data Centers: Architectures and techniques"}],"uid":"27466","created_gmt":"2011-04-21 12:46:35","changed_gmt":"2016-10-08 01:54:50","author":"Dani Denton","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2011-05-04T11:00:00-04:00","event_time_end":"2011-05-04T13:00:00-04:00","event_time_end_last":"2011-05-04T13:00:00-04:00","gmt_time_start":"2011-05-04 15:00:00","gmt_time_end":"2011-05-04 17:00:00","gmt_time_end_last":"2011-05-04 17:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"47223","name":"College of Computing"},{"id":"50875","name":"School of Computer Science"}],"categories":[],"keywords":[{"id":"12876","name":"Gong Zhang - Ph.D. Dissertation Defense Announcement"}],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1791","name":"Student sponsored"}],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}