國立高雄大學圖資館 |

Language: English

Back

A middleware architecture for servic...

Panahi, Mark.

A middleware architecture for service-oriented systems in support of accountability and predictability.

Record Type:	Electronic resources : Monograph/item
Title/Author:	A middleware architecture for service-oriented systems in support of accountability and predictability.
Author:	Panahi, Mark.
Description:	149 p.
Notes:	Source: Dissertation Abstracts International, Volume: 71-04, Section: B, page: 2575.
Notes:	Adviser: Kwei-Jay Lin.
Contained By:	Dissertation Abstracts International71-04B.
Subject:	Engineering, Computer.
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3397284
ISBN:	9781109690118

A middleware architecture for service-oriented systems in support of accountability and predictability.
Panahi, Mark.

A middleware architecture for service-oriented systems in support of accountability and predictability. - 149 p.

Source: Dissertation Abstracts International, Volume: 71-04, Section: B, page: 2575.

Thesis (Ph.D.)--University of California, Irvine, 2010.

Service-oriented architectures (SOA) provide enterprises the flexibility and cost-savings of dynamically composing business workflows from reusable services available anywhere remotely, potentially from third-party vendors. However, despite these new advantages, two aspects of SOA still remain a barrier to achieving end-to-end quality of service (QoS) guarantees for workflows: (1) in such a distributed setting, workflow behaviors are generally opaque such that it becomes difficult to identify the root causes of faults when they occur; and (2) current SOA frameworks do not provide the policies and mechanisms for deadline-driven workflow planning. Therefore, both enterprises and service vendors will need new tools to ensure the end-to-end QoS of workflows. To address opacity, enterprises will demand service accountability and transparency from service vendors. Likewise, service vendors will require the means to guarantee timely and predictable performance of their services.

ISBN: 9781109690118Subjects--Topical Terms:

384375
Engineering, Computer.

A middleware architecture for service-oriented systems in support of accountability and predictability.
LDR:03204nmm 2200325 4500 001 280797
005 20110119094956.5
008 110301s2010 ||||||||||||||||| ||eng d
020 $a 9781109690118
035 $a (UMI)AAI3397284
035 $a AAI3397284
040 $a UMI $c UMI
100 1 $a Panahi, Mark. $3 492912
245 1 2 $a A middleware architecture for service-oriented systems in support of accountability and predictability.
300 $a 149 p.
500 $a Source: Dissertation Abstracts International, Volume: 71-04, Section: B, page: 2575.
500 $a Adviser: Kwei-Jay Lin.
502 $a Thesis (Ph.D.)--University of California, Irvine, 2010.
520 $a Service-oriented architectures (SOA) provide enterprises the flexibility and cost-savings of dynamically composing business workflows from reusable services available anywhere remotely, potentially from third-party vendors. However, despite these new advantages, two aspects of SOA still remain a barrier to achieving end-to-end quality of service (QoS) guarantees for workflows: (1) in such a distributed setting, workflow behaviors are generally opaque such that it becomes difficult to identify the root causes of faults when they occur; and (2) current SOA frameworks do not provide the policies and mechanisms for deadline-driven workflow planning. Therefore, both enterprises and service vendors will need new tools to ensure the end-to-end QoS of workflows. To address opacity, enterprises will demand service accountability and transparency from service vendors. Likewise, service vendors will require the means to guarantee timely and predictable performance of their services.
520 $a To address these new requirements, this dissertation discusses how the Llama SOA middleware framework achieves accountable and predictable SOA both at the service-level, by enhancing existing open-source SOA middleware with monitoring, reconfiguration, and service execution reservation capabilities; and at the end-to-end workflow-level, by creating a distributed component infrastructure for problem detection, diagnosis, and deadline-based workflow composition.
520 $a Our experimental results indicate that using Llama's accountability infrastructure contributes only a modest amount of system overhead, and that the diagnosis process for pinpointing the root cause of process problems is swift and sufficiently accurate. Moreover, our evaluation of Llama's predictability framework demonstrates that service-level reservation management data structures are efficient with logarithmic time complexity over an indefinite time horizon, that a reserved service successfully executes on-time upon the arrival of its request, and that workflow reservations themselves can be made quickly and predictably by using a novel approach for achieving fast concurrent service reservations.
590 $a School code: 0030.
650 4 $a Engineering, Computer. $3 384375
650 4 $a Computer Science. $3 212513
690 $a 0464
690 $a 0984
710 2 $a University of California, Irvine. $b Electrical and Computer Engineering - Ph.D. $3 492898
773 0 $t Dissertation Abstracts International $g 71-04B.
790 1 0 $a Lin, Kwei-Jay, $e advisor
790 1 0 $a Chou, Pai $e committee member
790 1 0 $a Doemer, Rainer $e committee member
790 $a 0030
791 $a Ph.D.
792 $a 2010
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3397284