Current business critical environments increasingly rely on SOA standards to execute business operations. These operations are frequently based on Web service compositions that use several Web services over the internet and have to fulfill specific timing constraints. In these environments, an operation that does not conclude in due time may have a high cost as it can easily turn into service abandonment with financial and prestige losses to the service provider. In fact, at certain points, carrying on with the execution of an operation may be useless as a timely response will be impossible to obtain. This paper proposes a time-aware programming model for Web services that provides transparent timing failure detection. The paper illustrates the proposed model using a set of services specified by the TPC-App performance benchmark.
[1]
Flaviu Cristian,et al.
The Timed Asynchronous Distributed System Model
,
1998,
IEEE Trans. Parallel Distributed Syst..
[2]
Antonio Casimiro,et al.
The Timely Computing Base Model and Architecture
,
2002,
IEEE Trans. Computers.
[3]
Sam Toueg,et al.
Unreliable failure detectors for reliable distributed systems
,
1996,
JACM.
[4]
Mehmet Aksit,et al.
ECOOP'97 - Object-Oriented Programming : 11th European Conference Jyväskylä, Finland, June 9–13, 1997 : Proceedings
,
1997
.
[5]
Nancy A. Lynch,et al.
Consensus in the presence of partial synchrony
,
1988,
JACM.
[6]
Gregor Kiczales,et al.
Aspect-oriented programming
,
2001,
ESEC/FSE-9.
[7]
Cristina V. Lopes,et al.
Aspect-oriented programming
,
1999,
ECOOP Workshops.