- Briefing and discussion on network-centric operation and network enabled capability.
- Form a US &UK (or UK & US) Committee on SOA Engineering Collaboration for Network-Centric Operation.
- Group discussions and presentation.
- Plan a collaboration research plan.
- Develop joint statements.
US Representative
Dr. Raymond Paul
DoD OASD NII
Raymond.Paul@osd.mil
UK Representative
Prof. Jie Xu
University of Leeds
scsjx@leeds.ac.uk
Participants and their talk title:
UK (5)
¡°Service-Oriented Integration of Systems for Military Capability¡±
Duncan Russell
University of Leeds
¡°Dependable Architecture¡±
Jie Xu
University of Leeds
¡°CWID (Coalition Warrior Interoperability Demonstration) and BAE Systems approach to NEC¡± with Demonstration including the use of SOAs in UK Military Systems
John Davies
BAE Systems
Lu Liu
University of Leeds
David Webster
University of Leeds
US Government people:
¡°Human Interoperability Enterprise¡±
Alenka Brown
OSD/NII
Cam Camerron
SPAWAR
James Delambarlo
DARPA
¡°Dynamic Service-Oriented Engineering¡±
Raymond Paul
OSD/NII
Glen Stetller
SPAWAR/ Navy Department
Ken Turner
OSD/NII
US contractors/researchers:
¡°Integrated Dependability Modeling and Assessment of Net-Centric Systems¡±
Farokh Bastani????
University of Texas at Dallas
¡°Real-Time SOA¡±
Yann-Hang Lee
Arizona State University
Krishna Kavi
University of North Texas
¡°Developing dependable systems by maximizing component diversity
and fault tolerance¡±
Jeff Tian
Southern Methodist University
¡°Dependable Dynamic Composition for Service-Oriented Applications¡±
W. T. Tsai
Arizona State University
¡°Software as a Service¡±
Jeff Voas
¡°Addressing Service-Oriented Architecture (SOA) as a Complex Adaptive System¡±
George Hurlburt
Retired US Navy
Workshop Registration Fee
It has been estimated that the registration will be $40 to cover the coffee and lunch breaks.
Workshop participants plan to go out for dinner together after the first day of workshop.
Hotel arrangement:
Dial 1-703-416-4100 or 1-866-999-8439 (toll free)
Hotel is: Doubletree Hotel Crystal City-National Airport
Please mention the "10th IEEE CEC/EEE Conference"
Room rates are $154.00 (single) and $174.00 (double)
300 Army Navy Drive
Arlington, Virginia, United States 22202-2891
*Tel*: 1-703-416-4100
*Fax*: 1-703-416-4126
Click here for more information & reservations
<http://doubletree.hilton.com/en/dt/groups/personalized/DCAAEDT-IEE-20080720/index.jhtml>
Room rates:
*Single*: $154.00
*Double*: $174.00
These rates are exclusive of applicable sales/room tax, currently 10.25
percent. The special rates will apply at least three (3) days prior to and
three (3) days after the meeting dates based on availability.
Tentative Schedule
July 24 |
Speaker |
Topic |
|
8:30-8:50 |
Ray Paul |
Welcome and Dynamic Service-Oriented Engineering |
|
8:50-9:30 |
Alenka Brown |
Human Interoperability Enterprise |
|
9:30-10:10 |
Jie Xu |
Dependable Architecture |
|
10:10-10:40 |
Duncan Russell |
Service-Oriented Integration of Systems for Military Capability |
|
11:40-11:00 |
Break |
||
11:00-11:30 |
George Hurlburt |
Addressing Service-Oriented Architecture (SOA) as a Complex Adaptive System |
|
11:30-12:00 |
W. T. Tsai |
Dependable Dynamic Composition for Service-Oriented Applications |
|
12:00-12:30 |
Jeff Voas |
Software as a Service |
|
12:30-13:30 |
Lunch Break |
||
13:30-14:00 |
John Davies |
CWID (Coalition Warrior Interoperability Demonstration) and BAE Systems approach to NEC |
|
14:00-14:30 |
Y.H.Lee |
Real-Time SOA |
|
14:30-15:00 |
Jeff Tian |
Developing Dependable Systems by Maximizing Component Diversity |
|
15:00-15:30 |
Farokh Bastani |
Integrated Dependability Modeling and Assessment of Net-Centric Systems |
|
15:30-16:00 |
Break |
||
16:00-17:00 |
Planning and Discussion |
||
July 25 |
Speaker |
Topic |
|
8:30-8:45 |
Summary and Planning |
||
8:45-10:00 |
Group Discussion |
||
10:00-10:45 |
Group Presentation |
||
11:00-12:00 |
Joint Statement Preparation and Collaboration Plan |
||
1200- |
Lunch |
||
Abstract
---------------------------------------------------------------------------------------------------------------------
Title: Service-Oriented Integration of Systems for Military Capability
Dr Duncan Russell, School of Computing, University of Leeds, UK
---------------------------------------------------------------------------------------------------------------------
Abstract
Service oriented architecture (SOA) is becoming established in computing as
a means to integrate processing and data across organisations. System-level
integration can benefit from service oriented architectural descriptions and
loose coupling between the problem domain requirements and different system
solutions. An example problem domain is military capability. In particular,
Network Enabled Capability (NEC) from the UK Ministry of Defence (MoD).
Military capability can be described in terms of processes, which in turn
are sequences of functions that can be described as services. By decoupling
the service definition from the system solution, then different types of
system solutions can implement the described services
This presentation provides an overview of conceptual SOA and in the context
of military capability compares three levels of service integration:
business services, systems services and computing services. Secondly, the
presentation describes a framework for evaluating the performance and
effectiveness of service integration to compare different solutions in
delivering military capability.
------------------------------------------------------------------------------------------------------------------------
Title: Demonstrations including the use of SOAs in UK Military Systems
Prof John Davies, Insyte, BAE Systems, UK
------------------------------------------------------------------------------------------------------------------------
Abstract
This presentation will cover work done by BAE Systems UK, Insyte and other
organisations for CWID 07 and CWID 08 in Service Oriented Architectures and
related areas. These include:
* Bowman Tactical Networks
* Falcon Battlefield Communications
* Deployed Distributed Information Infrastructure
* Joint Command and Control Support Programme
* Digital Data Depository - Strategic to Tactical Multi Intelligence Hub
* Operational Logistic Information Systems
------------------------------------------------------------------------------------------------------------------------
Title:Architectural Support for Dependable System Evolution
Jie Xu
------------------------------------------------------------------------------------------------------------------------
Abstract
Building complex but dependable system with an ability to evolve in a controlled and dependable manner represents a major challenge in both system science and engineering. To address this challenge, we need to create an innovative change in the scientific foundation and engineering practice. The scientific foundation is needed for us to be able to build such systems whose dependability can be justified and guaranteed through their life time, even in a highly dynamic and open environment as well as in the presence of the most extreme threats.
Scientific foundation offers rigorous evidence which is needed to back up any claims about a system?s current and future service while sound engineering approaches (including techniques, measures and mechanisms) are required to actually deliver the service. Dependable system evolution is associated with at least two crucial properties of any system architecture: the system has well defined dependability attributes, and it has the ability to evolve through its life time in response to changes in its requirements, technology, and environments, without compromising its dependability.
My talk will start with basic concepts and models for dependable system evolution. I will discuss several techniques for design for dependable evolution? based on the concept of architectural options. I will then present further a service-oriented middleware system, called CROWN-C, with specific enhancements to support the development and assessment of highly dependable Grid and distributed applications.
------------------------------------------------------------------------------------------------------------------------
Title:Developing Dependable Systems by Maximizing Component Diversity and Fault Tolerance
Jeff Tian, Ph.D., P.E. Director, Software Engineering Program
Associate Professor, Dept of Computer Science and Engineering
Southern Methodist University, Dallas, Texas 75275
214-768-2861 tian@engr.smu.edu http://www.engr.smu.edu/~tian
------------------------------------------------------------------------------------------------------------------------
Abstract
For systems made up of different components to be dependable,they have to be composed in specific ways and the components
themselves must demonstrate some desirable properties that
contribute to overall system dependability.
In this presentation,
we focus on component diversity as a factor that affect the
resulting system dependability.
As a general rule, we would like to maximize component diversity so that
they can complement one another under unanticipated and
dynamically changing environment.
To do this,
we need a systematic way to assess the strengths and weaknesses
of individual components,
their ability to satisfy user requirement and advance customer values,
as well as the collective diversity,
and then perform system composition based on this assessment results.
A multi-dimensional evaluation scheme is proposed to assess
the various different component dependability attributes
and sub-attributes for classes of operational scenarios
and associated likelihood.
This evaluation scheme forms a matrix with dependability
attributes and its internal contributors
as one dimension and operational scenarios
as a second dimension.
For a fixed target environment,
the dependability attributes form a vector for each components.
Selection among components are based on maximization
of the efficiency frontier using data envelopment analysis (DEA).
If trade-offs are possible among the dependability attributes,
a single objective function can sometimes be formulated based on
a customer value assessment and quantification to select the ``best''
component for a specific setting.
In the case that direct measurement of a dependability attribute
is not available or not feasible,
an internal contributor to dependability attribute conversion
algorithm will be used to estimate the latter based on the
former and the environmental profile.
For unanticipated and changing target environment,
we are not only maximizing component diversity with respect
to dependability attributes (and sub-attributes) and customer values
attached to them,
but also diversity in dealing with different environments.Again, DEA and value-based objective functions can be used to make
this overall evaluation.
For experimental validation of our proposed selection procedure,
we plan to use a testbed based on fault injection and security
threat simulation.
Examples using elements of this approach in real projects are
included to illustrate the practical applicability of this approach.
------------------------------------------------------------------------------------------------------------------------
Title:High-Assurance Integrated Trustworthiness Assessment of Embedded Net-Centric Systems
F.B. Bastani, R.A. Paul, and I.-L. Yen
------------------------------------------------------------------------------------------------------------------------
Abstract
Applications of real-time embedded net-centric systems range from defense systems to emergency response systems, biomedical systems, transportation systems, etc. These are highly safety-critical and/or mission-critical systems where any failures can potentially have catastrophic consequences. Hence, there is an urgent need for high-confidence trustworthiness certification techniques for these large-scale complex distributed systems. The high-assurance certification must address several interrelated issues. For example, these systems must be highly reliable, available, and secure since they are often used in harsh environments that are susceptible to hardware failures and deliberate security attacks. They must also meet stringent temporal and spatial performance constraints, including being able to satisfy real-time processing and communication constraints and operate effectively under severe power, weight, and memory constraints.
All these complexities lead to great difficulties in achieving high-assurance certification for these critical systems. To address these problems, we present a three step certification process based on an evolving ontology of trustworthiness aspects. The first step is the identification and acquisition of various types of evidences for each basic component in the system that can be used to assess one or more trustworthiness aspects of the system. This provides a 2-dimensional set of basic trustworthiness data. The next step is the compositional assessment of each system-level trustworthiness aspect from the corresponding component-level trustworthiness aspect. This finally leads to the integrated trustworthiness assessment of the overall system, with one formulation for a specific mission and another for a given set of potential missions. The latter can be used to certify a set of strategic assets that can be dynamically composed to accomplish different tasks at different times.
The integrated trustworthiness assessment uses a combination of formal, statistical, and AI reasoning techniques to achieve rapid high-confidence certification of the trustworthiness of the overall system. These can be integrated into an environment to support high-confidence real-time, in-the-field trustworthiness certification techniques to dynamically rank multiple alternative designs for a given mission and enable the real-time selection and composition of the corresponding optimal system.
------------------------------------------------------------------------------------------------------------------------
Title:Addressing Service-Oriented Architecture (SOA) as a Complex Adaptive System
George Hurlburt
------------------------------------------------------------------------------------------------------------------------
Abstract
Service Oriented Architectures (SOA) offer great promise as they move systems integration away from a dependency on proprietary, often monolithic interfaces. A SOA uses services that are built on open standards. Moreover, services are defined in the context of specific domain-level operations. Thus, integration now becomes process and event centric rather than mandated by divergent legacy system dynamics. With the presumed flexibility and ease of SOA implementation also come a perplexing paradox.? The ease of service creation fosters a burgeoning growth of services. Each nuanced service connection, driven by domain requirements, often results in brand new, highly- specified service definitions. Other services are created that are so generic that they invoke very integration problems SOA is meant to resolve. Regrettably, this paradox has placed a premium on hard mappings, fixed data and structured functional definitions. The critical element of an SOA is whether services can be made discrete enough to be relevant, but generic enough to be manageable. This critical element can most easily be seen in the challenges confronting service discovery and composition in a net-centric world.
Current conventions suggest that service discovery occurs with robust meta-data or tagged descriptions of services and how those services function.? This in turn, creates another paradox. The meta-data tagging and highly structured definition of services is typically mandated to fit a specific vocabulary or standard. Unfortunately, services will be lost in the common case of exceptions, outliers or services with definitions that require a level of nuance that cannot be supported by the vocabulary. This drastically reduces the flexibility promise of SOA. On the other hand, leaving service definition to reflect the perspective of each individual developer and each specific implementation will create an unwieldy and perhaps destructive level of chaos in the dynamic System of Systems (SoS) environment.
These SOA paradoxes involve definition and meaning. SOA mimics the complexity of human communication and interaction. Recent advances in understanding complexity have made significant inroads into modeling human behavior. Complexity science posits that complex systems, whether based on human semantics or SOA discovery as a direct semantic derivative, can only be understood as dynamic networks of units that function with massive parallelism. Adaptation occurs in response to environmental stimuli. The overall system behavior is the result of a huge number of moment by moment interactions among the individual units. Thus, in complex systems, order is emergent as opposed to predetermined. The system's history is irreversible, and the system's future is often unpredictable. ?Likewise, the discovery of the correct combination of services to compose a desired transaction is naturally indeterminate and non-linear. Rigidly defined services have no utility in such an n-dimensional dynamic.
Recently the principles of complexity have been introduced in computer software design. The result is complex adaptive architectures that have been found to be especially applicable in the area of semantics and meaning discernment. Thus, one would expect their applicability to SOA and, in particular, service discovery and composition. As a result, it appears that service management built using complexity theory and adopting a complex adaptive architecture is a viable framework for research issues to resolve the following serious impediments to effective SOA implementation:
- Semantic complexity resulting from varying service definitions and interpretation
- Management? complexity? resulting from shifting service definitions
- Discovery complexity resulting from contextualized service and service composition definition
------------------------------------------------------------------------------------------------------------------------
Title:Real-time SOA
Yann-Hang Lee, Wei-Tek Tsai, and Yinong Chen
------------------------------------------------------------------------------------------------------------------------
------------------------------------------------------------------------------------------------------------------------
Title:Dependable Service-Oriented Composition
Wei-Tek Tsai
------------------------------------------------------------------------------------------------------------------------
Abstract
This talk will present new approaches to compose dependable service-oriented applications rapidly, and its implications to SOA system engineering.
------------------------------------------------------------------------------------------------------------------------
Workshop Address
2461 South Clark Street, Suite 560
Century Building 2
Crystal City, Virginia
Map
