國立虎尾科技大學 |

System, Architectural and Application Level Analysis of Big Data Applications for Performance and Energy-efficiency.

紀錄類型:	書目-語言資料,印刷品 : Monograph/item
正題名/作者:	System, Architectural and Application Level Analysis of Big Data Applications for Performance and Energy-efficiency./
作者:	Malik, Maria.
出版者:	Ann Arbor : ProQuest Dissertations & Theses, : 2018,
面頁冊數:	177 p.
附註:	Source: Dissertation Abstracts International, Volume: 79-11(E), Section: B.
Contained By:	Dissertation Abstracts International79-11B(E).
標題:	Computer engineering. -
電子資源:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10785775
ISBN:	9780438109469

System, Architectural and Application Level Analysis of Big Data Applications for Performance and Energy-efficiency.
Malik, Maria.

System, Architectural and Application Level Analysis of Big Data Applications for Performance and Energy-efficiency. - Ann Arbor : ProQuest Dissertations & Theses, 2018 - 177 p.

Source: Dissertation Abstracts International, Volume: 79-11(E), Section: B.

Thesis (Ph.D.)--George Mason University, 2018.

The volume of available data has exploded in recent years as a result of new social behaviors, societal transformations as well as advances in various branches of technology -- data sensing, data communication, data computation, and data storage. The world of big data is changing constantly that creates challenges to process the applications using existing solutions. Big data applications require computing resources and storage subsystems that can scale to manage massive amounts of diverse data. Furthermore, physical design constraints, such as power and density, have become the dominant limiting factor for scaling out servers. Therefore recent work advocates the use of low-power embedded cores in servers to address these challenges.

ISBN: 9780438109469Subjects--Topical Terms:

569006
Computer engineering.

System, Architectural and Application Level Analysis of Big Data Applications for Performance and Energy-efficiency.
LDR:04545nam a2200313 4500 001 931651
005 20190716101634.5
008 190815s2018 ||||||||||||||||| ||eng d
020 $a 9780438109469
035 $a (MiAaPQ)AAI10785775
035 $a (MiAaPQ)gmu:11668
035 $a AAI10785775
040 $a MiAaPQ $c MiAaPQ
100 1 $a Malik, Maria. $0 (orcid)0000-0001-8425-2501 $3 1213847
245 1 0 $a System, Architectural and Application Level Analysis of Big Data Applications for Performance and Energy-efficiency.
260 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2018
300 $a 177 p.
500 $a Source: Dissertation Abstracts International, Volume: 79-11(E), Section: B.
500 $a Adviser: Houman Homayoun.
502 $a Thesis (Ph.D.)--George Mason University, 2018.
520 $a The volume of available data has exploded in recent years as a result of new social behaviors, societal transformations as well as advances in various branches of technology -- data sensing, data communication, data computation, and data storage. The world of big data is changing constantly that creates challenges to process the applications using existing solutions. Big data applications require computing resources and storage subsystems that can scale to manage massive amounts of diverse data. Furthermore, physical design constraints, such as power and density, have become the dominant limiting factor for scaling out servers. Therefore recent work advocates the use of low-power embedded cores in servers to address these challenges.
520 $a In this research, first through comprehensive system and micro-architectural level analysis, we characterize the big data applications on big Xeon and little Atom-based server architecture to demonstrate how the choice of big vs little core-based server for energy- efficiency is influenced by the size of data, performance constraints, and presence of accelerator. Second, we analyze the performance, energy efficiency and cost efficiency of MapReduce on big and little cores across a large range of tuning parameters at application, system and architecture levels. The goal is to identify the right computing platform for Big Data analytics processing that can provide a balance between processing capacity, cost efficiency, and energy efficiency. This characterization analysis helps guiding scheduling decision in future cloud-computing environment equipped with heterogeneous server architectures. Third, given that the choice of big vs little core can be impacted by various tuning knobs, we also study the impact of application, system and architecture parameters and the interplay among them on performance and energy-efficiency on MapReduce phases of big data applications.
520 $a With the significant increase in the volume of data, more applications are migrating to cloud. Thus, datacenters become the computer platforms of choice to process diverse applications in the emerging domain of big data. However, cloud/datacenter infrastructure does not scale as fast as the input data volume and computational requirements of big data and analytics technologies. Thus, more applications need to share CPU at the node level that could have large impact on performance and operational cost. To address this challenge, we show that, concurrently fine-tune parameters at the application, microarchitecture, and system levels create opportunities to co-locate applications at the node level and improve energy- efficiency of the server while maintaining performance. Co-locating and self-tuning of unknown applications are challenging problems, especially when co-locating multiple big data applications concurrently with many tuning knobs, potentially require exhaustive brute-force search to find the right settings. This research challenge upsurges an imminent need to develop a technique that co-locates applications at a node level and predict the optimal system, architecture and application level configure parameters to achieve the maximum energy efficiency. Towards this goal, we develop an Energy-Efficient Co-Locating and Self-Tuning (ECoST) technique for co-located MapReduce applications to enhance their energy-efficiency. ECoST collects run-time hardware performance counter data and implements various machine learning based models to predict the energy- efficiency of co-located applications.
590 $a School code: 0883.
650 4 $a Computer engineering. $3 569006
690 $a 0464
710 2 $a George Mason University. $b Electrical and Computer Engineering. $3 1187212
773 0 $t Dissertation Abstracts International $g 79-11B(E).
790 $a 0883
791 $a Ph.D.
792 $a 2018
793 $a English
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10785775