Overview: This project investigates the quality and robustness
of service discovery protocols in distributed systems. Specifically,
the project focuses on the response of service discovery protocols
to dynamic change under hostile and volatile conditions, such as
may be caused by network process, node, and link failures. The goals
of the project are to test the behavior and resilience of dynamic
service discovery protocol specifications and to compare and contrast
different service discovery system designs.
Industry Need Addressed: The design of future commercial
distributed software systems that employ service discovery in uncertain,
dynamic environments will demand new analysis approaches and tools.
Numerous industry trends foretell a future where software components
will need to discover other network services and applications in
a distributed environment characterized by dynamic modification,
recomposition, and change in topology. Recently a number of commercial
Service Discovery Protocols have emerged that enable dynamic elements
in a network to discover each other and to collaborate in order
to meet an application need. To assist in developing new tools and
approaches to the design and analysis of such systems, we investigate
the use of architectural modeling techniques to improve analysis
of specifications.
NIST/ITL Approach: NIST/ITL is working with interested parties
in industry to analyze critical aspects of service discovery protocols
in a series of studies. The approach used in these studies entails
constructing architectural models of selected aspects of discovery
protocol specifications. Relevant properties are specified for the
protocols to satisfy in order to demonstrate a desirable quality
of service, and metrics are defined for comparing the model behavior.
Using the properties and metrics as a basis, the models are then
analyzed to yield information about the utility of service discovery
protocol designs under dynamic conditions.
Impact: NIST/ITL, through a series of publications describing
the results of these studies, provides critical information to developers
of commercial service discovery protocols about new approaches to
improving the quality of their specifications using architectural
modeling techniques. In the future, we plan to expand this study
to encompass quality and robustness of large-scale grid computing
systems under hostile and volatile conditions..