國立虎尾科技大學 |

Cloud-Based Log Store Backed Streaming Data Management.

紀錄類型:	書目-語言資料,手稿 : Monograph/item
正題名/作者:	Cloud-Based Log Store Backed Streaming Data Management./
作者:	Pathirage, Milinda.
面頁冊數:	1 online resource (129 pages)
附註:	Source: Dissertation Abstracts International, Volume: 79-09(E), Section: B.
Contained By:	Dissertation Abstracts International79-09B(E).
標題:	Computer science. -
電子資源:	click for full text (PQDT)
ISBN:	9780355947427

Cloud-Based Log Store Backed Streaming Data Management.
Pathirage, Milinda.

Cloud-Based Log Store Backed Streaming Data Management. - 1 online resource (129 pages)

Source: Dissertation Abstracts International, Volume: 79-09(E), Section: B.

Thesis (Ph.D.)--Indiana University, 2018.

Includes bibliographical references

The data-driven economy increasingly requires companies to draw in and analyze data in real time from numerous sources. Cloud infrastructure is an attractive approach as it is highly responsive to on-demand scaling needs. In parallel, a variant of distributed stream processing systems, called log store backed stream processing systems, have emerged that support both streaming queries and historical queries on real-time sources.

Electronic reproduction.
Ann Arbor, Mich. :
ProQuest,
2018

Mode of access: World Wide Web

ISBN: 9780355947427Subjects--Topical Terms:

573171
Computer science.
Index Terms--Genre/Form:

554714
Electronic books.

Cloud-Based Log Store Backed Streaming Data Management.
LDR:03355ntm a2200361Ki 4500 001 916687
005 20180927111922.5
006 m o u
007 cr mn||||a|a||
008 190606s2018 xx obm 000 0 eng d
020 $a 9780355947427
035 $a (MiAaPQ)AAI10808720
035 $a (MiAaPQ)indiana:15166
035 $a AAI10808720
040 $a MiAaPQ $b eng $c MiAaPQ $d NTU
100 1 $a Pathirage, Milinda. $3 1190495
245 1 0 $a Cloud-Based Log Store Backed Streaming Data Management.
264 0 $c 2018
300 $a 1 online resource (129 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertation Abstracts International, Volume: 79-09(E), Section: B.
500 $a Adviser: Beth Plale.
502 $a Thesis (Ph.D.)--Indiana University, 2018.
504 $a Includes bibliographical references
520 $a The data-driven economy increasingly requires companies to draw in and analyze data in real time from numerous sources. Cloud infrastructure is an attractive approach as it is highly responsive to on-demand scaling needs. In parallel, a variant of distributed stream processing systems, called log store backed stream processing systems, have emerged that support both streaming queries and historical queries on real-time sources.
520 $a This thesis contributes optimizations and abstractions to improve the efficiency and ease of use of log store-backed stream processing in clusters of cloud resources that use popular (non-specialized) cloud resources. Challenges for moving log store backed stream processing infrastructures to the cloud include high turnaround times of programming abstractions that requires compiling, deriving efficient and cost-effective cloud infrastructure configurations for a given fast data workload, and evaluation of different infrastructure choices.
520 $a The specific contributions are several. First, a minimal set of extensions to standard SQL to bring low barrier and the precise semantics of SQL to queries that execute simultaneously over data streams (in movement) and tables (at rest as a changelog). We prototype the proposed language in SamzaSQL that brings interactive querying capabilities to log store backed stream processing. Second, a resource requirement estimation model and a vector bin packing approach for deriving cost-effective log store deployments in the cloud. Proposed approach incorporates a machine learning model for dynamically adjusting the per disk I/O capacity to avoid oversubscription of disk resources.
520 $a Finally, an extensible software framework for developing workload generators and benchmarks that target fast data systems, especially log stores and log store backed stream processing systems. Proposed software framework enables modeling, comparison and validation of fast data systems and infrastructures by providing a programming abstraction that facilitates distributed execution and metrics collection in addition to built-in workload generators and benchmarks.
533 $a Electronic reproduction. $b Ann Arbor, Mich. : $c ProQuest, $d 2018
538 $a Mode of access: World Wide Web
650 4 $a Computer science. $3 573171
655 7 $a Electronic books. $2 local $3 554714
690 $a 0984
710 2 $a ProQuest Information and Learning Co. $3 1178819
710 2 $a Indiana University. $b Computer Sciences. $3 1179305
773 0 $t Dissertation Abstracts International $g 79-09B(E).
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=10808720 $z click for full text (PQDT)