One of MC jaeger's papers, "QoS aggregation for Web service composition using workflow patterns" (Edo

Michael C. Jaeger, Gregor rojec-Goldmann, Gero M ?, "QoS aggregation for Web service composition using workflow patterns, "Edoc, pp.149-159, Enterprise Distributed Object Computing conference, eighth IEEE International (edoc '04), 2004

Time: 2 hours

recently I was reading some Code . I was a little worried when I didn't read my paper for a few days. because I have no time to read new papers, I have reviewed several previous papers.
Michael C. jaeger (Berlin University of Technology)
the research direction is quality-of-service and distributed systems ontologies, semantic web and Semantic Web Services.
In, several papers on QoS aggregation of composite services based on Workflow structure were cited for high usage.
QoS aggregation for Web service composition using workflow patterns (GS: 83), edoc, 2004
QoS aggregation in Web Service compositions (GS: 41), eee, 2005
Addendum: QoS aggregation using WF patterns (2005)
optimising Quality-of-Service for the composition of electronic services . phD thesis, Berlin University of Technology, January 2007.
these papers are not difficult. There are no complicated algorithms , no complete system, and no experimental verification, the main reason is that some summative work is highly cited. The possible reason is that
(1) the publishing time is earlier (2) content about the basic issues in the field of composite services (3) the paper itself is easy to understand

1. this article involves the work of workflow Pattern Based on Van der Aalst, mainly discusses sequence, loop, XOR-SPLIT/XOR-JOIN (conditional selection), and-Split/and-join (concurrency ), and-Split/M/n-join (often used in Fault Tolerance mode), or-Split/or-join and or-Split/M/n-join.

2. The involved Qos attributes include execution time, cost, encryption, throughput, and uptime probability.
For execution time and cost, the author lists the conditions for getting the maximum and minimum values after aggregation.
For example, for execution time, in the sequence, loop, and-and structure, the maximum and minimum values of QoS for aggregation are the same; but in the XOR-XOR structure, the maximum and minimum values are determined by the execution time and the minimum service in the branch.
The uptime probability discussed in (s4.3) seems to be the reliability,Article"Is regarded as a decreasing dimension" should be written by mistake. This should be increasing dimension. In addition, the words and phrases in this article are rough in layout, misspelled, and misspelled.

3. When QoS aggregation is performed, this article has some assumptions (s4.2)
(1) Service independence. (The papers on considering QoS association between services include generation 06 and ye Shiyang 08)
(2) QoS claimed by the service is credible. (This is the research on Trust & reputation of web service QoS, which was first found in EM maximilien. Since then, there have been many researches in this field)
(3) Uniformity: the definition of web service QoS is usually different. Therefore, the consistent principle should be followed during the discussion.
(4) equipartition: for each branch, it is assumed that the probability of selection is equal (if historical data exists, more reliable probability data can be obtained)

4. The difference between the composition of Web Service and workflow is discussed in (S1 ).
(1) services in composition can be dynamically discovered, bound, and replaced
(2) There can be many redundant services in composition for selection.
(3) human activities are not supported in Web service composition (however, some Web Service combinations support human activities, such as bpel4people)

The QoS aggregation table in the paper is attached to facilitate search