Session Abstracts
Building Survivable Systems from COTS
Components: An Impossible Dream?
Moderator: Nancy Mead, Software Engineering Institute
Panelists: Joe Besselman, USAF; Nancy Mead and Howard Lipson,
Software Engineering Institute; and Jeff Voas, Cigital
Panel Discussion
Monday, February 4, 11:00 a.m. - 12:15 p.m.
Much of the literature on COTS-based systems and the license
agreements for COTS products concede that such systems are not
suitable for critical applications for business, government,
and defense. However, COTS-based systems are already being used
in domains where significant economic damage and loss of life
are possible. Can we ever build such systems so that the risks
associated with using COTS components in critical applications
are commensurate with those typically taken in other areas of
life and commerce?Building survivable systems using COTS components
is a daunting task because the design team typically has little
or no access to the artifacts of the software engineering process
used to create the components. These artifacts are the primary
sources from which assurance evidence for a composite system
is derived, and lack of direct access to the artifacts greatly
increases the risks of building critical applications. The key
question to be addressed is this: Can those who acquire, design,
implement, operate, maintain, or evolve systems that use COTS
components adequately manage the risks of using these components
in critical applications? This panel will debate if, when, and
how COTS components can be used to build survivable systems
and discuss strategies for risk mitigation.
Building Systems from Commercial Components
Robert Seacord, Software Engineering Institute
Monday, February 4, 3:00 p.m. - 5:00 p.m.
This tutorial describes fundamental ideas of component-based
development from building Systems from Commercial Components,
a book in the SEI Series in Software Engineering. Component
ensembles are introduced as a fundamental design abstraction.
Ensembles-sets of components that work together-expose component
dependencies and shift the emphasis from selecting individual
components to selecting sets of components. Blackboards are
introduced as a fundamental design notation. Blackboards depict
what is currently known about an ensemble and, just as important,
what remains to be discovered. Blackboards serve to document
a design and known areas of design risk. A risk-driven discovery
process called R3 (Risk, Realize, Repair) is introduced that
exposes design risk and defines ensemble feasibility criteria.
A prototyping process called model problems is also introduced
that generates situated component expertise and establishes
ensemble feasibility. These techniques are illustrated using
a case study taken from a large, information system modernization
effort.
This tutorial is intended for individuals who are participating
in a component-based development effort (e.g., system architects,
chief engineers, project managers, software engineers, programmers).
The Challenges of COTS-Centric Software
Development and Modernization: Some Experiences from the Field
Joe Besselman, U.S. Air Force
Presentation
Tuesday, February 5, 3:35 p.m. - 4:20 p.m.
The burgeoning Information Technology (IT) sector continues
to flood the marketplace with a plethora of commercial off-the-shelf
(COTS) software products performing a near endless variety of
functions. These products carry special appeal for today's IT
managers because they offer many seductive benefits over traditional,
organically developed software. In this era of increased global
competition in the commercial marketplace and flat government
budgets-particularly in the U.S. Department of Defense (DoD),
the world's single largest IT customer-efficiencies promised
by COTS vendors are aggressively pursued by IT managers. A COTS
product that is capable of replacing all or part of a piece
of organic software enables IT managers to either divert their
organic development labor to another task or reduce their organization's
labor pool. For the past six years, the US Air Force has aggressively
pursued COTS products not only to modernize its suite of several
hundred support systems, but also to integrate these applications.
In the course of these experiences, we have catalogued a series
of experiences, both good and bad, that we use to help chart
future modernization activities. This paper discusses important
technical and organizational considerations or "challenges"
facing an IT manager contemplating replacing an organic development
operation with a singular product or suite of COTS products:
scalability, volatility, maturity, business process viability,
interfaces, customization, maintenance and support, metadata
standardization/sharing, enterprise architecture integration,
legacy/internet revolution synergy, politics, bureaucracy, security,
and help desk support.
Combined Selection of COTS Components
Xavier Franch, Universitat Politècnica de Catalunya
Presentation
Wednesday, February 6, 1:30 p.m. - 2:15 p.m.
In this paper the problem of the combined selection of COTS
components is analyzed in organizations of a specialized nature.
This is currently a process of great interest: once many organizations
have implemented recently ERP systems for supporting their central
management areas, they need to select specialized components
for other more particular business areas. We propose a model
of combined selection of components that is divided into two
levels. The global level corresponds to combined selection and
includes initial planning, enacting individual selection processes,
proposing scenarios to evaluate, and final selection of COTS
components. In the local level we locate all the individual
selection processes of the different organizational areas under
the supervision of the global level. The model presented here
arises from the observation of a real ongoing case of collective
selection that takes place in a big hospital.
COTS Test Strategy
Moderator: Christina Laiacona, BASF Corporation
Panelists: John Dean, National Research Council; George Grigonis,
Computer Validation Group (CVIG); and Randall Rice, Rice Consulting
Services
Panel Discussion
Wednesday, February 6, 10:30 a.m. - 11:45 a.m.
One of the pressures created by COTS applications is the paradigm
shift for new testing strategies. This spirited panel discussion
will begin the dialogue on COTS Test Strategy for product evaluation,
deployment and maintenance. The following questions will be
addressed during the session:
- What are the testing considerations for a COTS Product Evaluation?
- What are the management processes and people considerations
of a testing strategy for COTS applications?
- Are there different approaches to testing COTS applications
according to the development risk of the systems' ability
to capture, create, store, manage, and
archive trustworthy data?
- What are the considerations for surveillance/maintenance
of COTS systems in a production environment?
COTS-Based System Engineering: The
Linguistics Approach
Nguyen Thanh Quang,
Department of Information and Decision Systems, ESSEC Business
School, France
Presentation
Wednesday, February 6, 4:25 p.m. - 5:10 p.m.
As software systems become more and more complex and software
artifacts developed by third party emerge frequently, the move
towards COTS-based system engineering is a natural maturation
process of software engineering. However, current requirements
engineering methods for COTS-based systems have proven to be
less successful than expected. The paper is an attempt to identify
key properties of existing COTS software and other software
components. We suggest a systematic approach which examines
COTS-related issues under three linguistic levels: lexicon,
syntax, and semantics. The driving force behind this approach
by analogy is to outline a new requirements engineering method
for COTS-based systems.
COTS-Based Systems (CBS) Functional Density:
A Heuristic for Better CBS Design
Chris Abts, USC Center for Software Engineering
Presentation
Tuesday, February 5, 3:35 p.m. - 4:20 p.m.
The conventional rationale for using COTS (commercial off-the-shelf)
components is that the more a software system is built from
COTS products, the lower the cost of initial development. Less
understood is that during the long term sustainment phase--from
deployment through retirement-the cost of maintaining a COTS-based
system generally increases as the number of COTS products used
increases. There is a tension between the imperative to maximize
the use of COTS components to ease CBS development yet minimize
the use of COTS components to ease CBS maintenance. A heuristic
called the "CBS Functional Density Rule" is proposed
to reconcile these two conflicting views. A corresponding metric
for characterizing the "efficiency" of a given CBS
design relative to the "COTS Functional Density" is
then suggested. The paper concludes with suggestions for additional
research to further validate the empirical foundations of the
proposed heuristic and associated metric.
COTS-Based Systems: Keys to Success
Patricia Oberndorf, Software Engineering Institute
Tutorial
Monday, February 4, 1:15 p.m. - 2:45 p.m.
This tutorial will introduce you to the challenges and opportunities
encountered when incorporating commercial off-the-shelf (COTS)
products into software-intensive systems. Attendees already
know how to cope with traditional development programs; this
tutorial emphasizes differences in working on COTS-based systems.
The presentation covers fundamental definitions, major actions
necessary for success with COTS-based systems, and then concludes
with recommendations. The goal is to heighten the attendee's
awareness about the business, contractual, and engineering issues
associated with the acquisition of COTS-based software-intensive
systems. The session will be highly interactive; questions will
be welcome, and the presenter will facilitate as much discussion
as time allows.
The Limitations of Current Decision-Making
Techniques in the Procurement
of COTS Software Components
Cornelius Ncube, Zayed University Dubai, United Arab Emirates
Presentation
Tuesday, February 5, 4:25 p.m. - 5:10 p.m.
One of the critical issues for the COTS-based development process
is COTS software assessment and decision-making. In order to
select or recommend a suitable required product, the evaluated
alternatives must be ranked according to their perceived relative
importance to meet the customer's requirements. Decision-making
techniques have been used for this purpose. However, most of
the current decision-making techniques available are not adequate
for COTS software evaluations and assessments due to their underlying
assumptions and judgement value systems. There is a need for
new requirements-driven decision-making techniques for the COTS-based
development paradigm. This paper proposes a flexible requirement-driven
decision-making process that recognizes the trade-offs between
requirements, decisions, and COTS software selection.
Definition and Classification of COTS:
A Proposal
Marco Torchiano, NTNU - IDI
Presentation
Tuesday, February 5, 4:25 p.m. - 5:10 p.m.
COTS-based development impacts several issues in software development.
New techniques have been proposed or existing ones have been
adapted. Several approaches have been proposed for effort and
size estimation, product selection, and architectural mismatches
identification. But before considering these issues, we must
clarify a fundamental question: What is a COTS product? According
to the literature, a COTS product seems to be anything from
an operating system to a UI widget. It appears obvious that
a finer level of granularity is required if we want to acquire
a deeper insight in COTS related issues. This paper proposes
a COTS classification scheme that is as inclusive as possible.
It is intended to provide both researchers and practitioners
with a tool for more precise characterization of their work.
The next research step will be to validate, first by speculation
and later by empirical study, the influence of COTS classes
on issues in COTS-based development.
Estimating COTS-Based Software Systems
Chris Abts, USC Center for Engineering and Betsy Clark, Software-Metrics
Tutorial
Monday, February 4, 1:30 p.m. - 5:00 p.m.
This will be a half-day tutorial (in 2 parts) on CBS estimation
in general and how to apply the COTS model in particular.Part
1 - general overview of CBS issues over the entire life cycle
from inception through retirement, especially items potentially
impacting cost and schedule. Included will be some discussion
of CBS "best practices" for mitigating project risks.Part
2 - detailed review of COCOTS and its relation to COCOMO II.
This will focus primarily on the COCOTS development model but
will also touch upon the backend maintenance model that has
been proposed data. Discussion will include 1) the level of
accuracy that can realistically be expected from the model at
this time, 2) use cases and scenarios illustrating how the model
might be used, and 3) a tool demonstration and exercise working
through an example of a CBS life cycle cost and schedule estimation.
European COTS User Working Group: Analysis
of the Common Problems and Current Practices of European COTS
Users
Sandy Tay, European Software Institute
Presentation
Tuesday, February 5, 4:25 p.m. - 5:10 p.m.
The use of commercial off-the-shelf (COTS) software is increasingly
becoming a necessity for many European organizations. But this
necessity introduces new problems and changes for these organizations
at all levels both from the business and technical points of
view. Funded by the Information Society Technologies Programme
of the European Commission, the European COTS User Working Group
(ECUA) was created to address common problems faced by the European
COTS users, achieve consensus on their solutions and produce
a favourable impact in the overall COTS market. This paper is
written based on the discussions of the special interest group
sessions on the COTS issues from the business and the information
technology (IT) perspectives at the first ECUA Workshop, held
at the European Software Institute in Zamudio, Spain, in which
40 participants from all over Europe (and 1 from Canada) participated.
Evaluating COTS Software Products
John Dean, National Research Council and Grace Lewis, Software
Engineering Institute
Tutorial
Monday, February 4, 1:15 p.m. - 5:00 p.m.
This tutorial addresses the process and techniques that can
be employed in the evaluation of commercial off-the-shelf (COTS)
products used to construct software-intensive systems. After
attending, practitioners will be able to understand the impact
of COTS products on the system development process, determine
evaluation requirements for COTS software, develop COTS software
evaluation criteria, select COTS software evaluation techniques,
and employ a COTS software evaluation process that addresses
the inherent tradeoffs. The presentation covers fundamental
definitions; an overview of some of the basic principles of
COTS product evaluation; a process framework for COTS software
product evaluation; and some useful techniques. We will describe
experiences and practices, based on real-world case studies,
for COTS product evaluation. The instructors, Grace Lewis and
John Dean, deliver interactive instruction, interspersed with
realistic exercises. They invite questions and discussion as
much as possible.
Five Hurdles to the Successful Adoption
of Component-Based COTS in
a Corporate Setting
Anthony Earl, Sun Microsystems Inc.
Presentation
Wednesday, February 6, 4:25 p.m. - 5:10 p.m.
In this paper, the author reports experiences with component-based
technologies that have been observed from the roles of component
technology engineer, evangelist, and consultant within the corporate
world. The goals are firstly to capture those observations for
the interest and benefit of component vendors and secondly to
offer some interest and basis for further studies that may show
whether or not such experiences are commonplace or not. Armed
with such data, there is clearly a better chance that component
vendors and corporate entities can form collaborations through
which each can achieve their aims.
Identifying Evolvability for Integration
Rose Gamble, University of Tulsa
Presentation
Wednesday, February 6, 1:30 p.m. - 2:15 p.m.
The seamless integration of commercial off-the-shelf (COTS)
components offers many benefits associated with reuse. Even
with successful composite applications, unexpected interoperability
conflicts can arise when COTS products are upgraded, new components
are needed, and the application requirements change. Recent
approaches to integration follow pattern-based design principles
to construct an integration architecture for the composite application.
This integration architecture provides a foundation for addressing
the problematic interactions among components. However, little
attention has been paid to the evolvability of these architectures
and their embedded functionality. In this paper, we discuss
the need for design traceability based on the history of interoperability
conflicts and resolution decisions that comprise an integration
architecture. Additionally, we advocate that certain functional
aspects of a pattern can be pinpointed to resolve a conflict.
Combining these two aspects of integration architecture design,
we illustrate that often evolution is possible with minimal
change to the integration solution.
The Integration of COTS/GOTS Within NASA's
HST Command and Control System
Thomas Pfarr, Computer Sciences Corporation
Presentation
Wednesday, February 6, 3:35 p.m. - 4:20 p.m.
NASA's mission critical Hubble Space Telescope (HST) command
and control system has been re-engineered with commercial off-the-shelf/government
off-the-shelf (COTS/GOTS) and minimal custom code. This paper
focuses on the design of this new HST Control Center System
(CCS) and the lessons learned throughout its development. CCS
currently utilizes more than 30 COTS/GOTS products with an additional
½ million lines of custom glueware code. The new CCS
exceeds the capabilities of the original system while significantly
reducing the lines of custom code by more than 50%. The lifecycle
of COTS/GOTS products will be examined including the package
selection process, evaluation process, and integration process.
The advantages, disadvantages, issues, concerns, and lessons
learned for integrating COTS/GOTS into NASA's mission critical
HST CCS will be examined in detail. This paper will reveal the
many hidden costs of COTS/GOTS solutions when compared to traditional
custom code development efforts; this paper will show the high
cost of COTS/GOTS solutions including training expenses, consulting
fees, and long-term maintenance expenses.
Issues in Developing Security Wrapper
Technology for COTS Software Products
John Dean, National Research Council Canada
Presentation
Wednesday, February 6, 2:20 p.m. - 3:05 p.m.
The use of commercial off-the-shelf (COTS) software products
as components of large-scale systems has become more and more
pervasive. Many of these products do not incorporate security
policies as an integral part of the application. One of the
interesting questions that has been asked by systems integrators
is "Can you build secure applications using insecure components?"
We have been investigating ways to protect data that is shared
between two or more independent, insecure COTS products. Our
experiments in implementing secure data storage and transfer
using such productshave been directed toward building data encryption
wrappers that interact with various COTS products. We have examined
the ability of such wrappers to adapt to updates or substitutions
of the COTS products. We describe the methods for building a
security wrapper, identify the problems encountered during the
development, and present interim results. We also describe the
effort expended in porting the application.
Lessons Learned Integrating COTS into
Systems
Tom Baker, The Boeing Company
Presentation
Tuesday, February 5, 2:20 p.m. - 3:05 p.m.
This paper presents lessons learned by the author over 15 years
of experience integrating COTS software into systems at The
Boeing Company. One key lesson has been to distinguish development
vs. customization vs. integration vs. configuration and understanding
the corresponding impacts on tools, architectures, and even
methodologies. Methodology impacts can destroy a project if
they aren't recognized and mitigated in a timely manner. The
paper walks through four COTS projects, discussing the challenges
of each and how they were overcome. The COTS integration project
involving a workflow engine provides an extreme example of methodology
impact.
Meeting the Challenges of Commercial Off-the-Shelf
(COTS) Products: The Information Technology Solutions Evolution
Process (ITSEP)
Cecilia Albert, Software Engineering Institute
Presentation
Wednesday, February 6, 3:35 p.m. - 4:20 p.m.
Government and private organizations are escalating their use
of commercial off-the-shelf (COTS) products in critical business
systems. These organizations find little success with the traditional
development approach-that is, the process of defining requirements,
formulating an architecture, and then trying to find COTS products
to meet the specified requirements within the defined architecture.
We describe an alternative approach, based on the Rational Unified
Process™ (RUP), that modifies the acquisition and development
processes to more effectively leverage the COTS marketplace
through concurrent discovery and negotiation of user needs and
business processes, applicable COTS technology and products,
the target architecture, and programmatic constraints.
Merging Integration Solutions for Architecture
and Security Mismatch
Rose Gamble, The University of Tulsa
Presentation
Tuesday, February 5, 2:20 p.m. - 3:05 p.m.
Integrating COTS products into a composite application can reduce
development effort and associated costs. A major drawback comes
from interoperability problems that hinder the seamless integration
of components. The two prominent types of problems are architecture
mismatch and security mismatch. Because of their distinct properties,
each problem is currently analyzed separately. The results are
integration solutions that are constructed in isolation. Combining
these solutions can yield another set of problems if their functionality
is conflicting, duplicated, or overly complex. It is imperative
to address these issues in component-based software development.
In this paper, we depict the architectural differences among
components, their security access control policies, and the
integration solutions that result from independent analysis.
This is the first step toward including architectural interoperability
issues and security conflicts in the design of an encompassing
solution for an integrated application. We show a composition
of the two solutions, highlighting redundancy and complexity.
On Building Testable Components
Jerry Gao, PhD San Jose State University
Presentation
Tuesday, February 5, 3:35 p.m. - 4:20 p.m.
Component engineering is gaining substantial interest in the
software engineering community. A lot of research efforts have
been devoted to analysis and design methods for component-based
software. However, few papers address the testing of software
components and component-based software. This paper focuses
on how to build testable software components by increasing the
testability of software components. It discusses the component
testability concept in terms of its factors, issues, and solutions
in design and testing of software components. Moreover, the
paper introduces the concept of the testable bean, and proposes
a new way to construct a testable bean based on a testable architecture
and well-defined built-in interfaces. In addition, the paper
also reports our efforts on developing a test bed to achieve
automation of test beans.
Once Burned, Forever Learned-Vendors Be
Warned
Moderator: Will Tracz, Software Engineering Institute
Panelists: Ronald Kohl, Titan Systems; Thomas Baker, The Boeing
Company; and Anthony Earl, Sun Microsystems
Panel Discussion
Tuesday, February 5, 10:30 a.m. - 11:45 a.m.
This goal of this panel is to generate a "wish list"
of items that COTS-based system developers would like COTS vendors
to better understand. Panelists will provide gap analysis based
personal experience and "war stories" to support their
individual wish lists. This will be followed by prioritization
by the panel members based on general discussion and inputs
from the panel session attendees.
Process Evolution in Large-Scale COTS
Reengineering Within the United States Department of Defense
James Smith, II, Software Engineering Institute
Presentation
Wednesday, February 6, 4:25 p.m. - 5:10 p.m.
This paper reports on empirical research into the organizational
and process changes necessary to implement large-scale reengineering
using COTS. The research goal is to identify how these changes
can be effected within the context of the existing United States
Department of Defense acquisition management, requirements management,
and planning, programming, and budgeting systems. While this
effort is still in its early stages, initial results indicate
that there are significant challenges across the board, especially
with developing and defending budget inputs to support a spiral
acquisition model.
A Process for COTS Software Product Evaluation
Patricia Oberndorf, Software Engineering Institute
Presentation
Wednesday, February 6, 2:20 p.m.-3:05 p.m.
The growing use of commercial products in large systems makes
evaluation and selection of appropriate products an increasingly
essential activity. However, many organizations struggle in
their attempts to select an appropriate product for use in systems.
As part of a cooperative effort, the Software Engineering Institute
(SEI) and the National Research Council Canada (NRC) have defined
a tailorable software product evaluation process that can support
organizations in making sound, carefully reasoned product decisions.
This paper describes that process.
Realizing the Potential for COTS Utilization:
A Work in Progress
Rhoda Shaller Hornstein, NASA Headquarters
Presentation
Tuesday, February 5, 1:30 p.m. - 2:15 p.m.
For over a decade, the U.S. Government has been emphasizing
its preference for using commercial off-the-shelf (COTS) products
as a way to reduce program costs and accelerate schedules. The
results of this initiative have been mixed, with many programs
reporting fewer benefits from COTS usage than its advocates
had forecast. This paper explores the reasons for the unfulfilled
potential of COTS utilization and presents some new considerations
for addressing the COTS challenges. Between 1992 and 1996, a
NASA task force called the COST LESS Team developed recommendations
for acombined technical architecture and electronic-commerce
marketplace strategy for reducing the cost and cycle time of
systems for space programs, while improving their quality and
responsiveness to customer needs. COTS buying was a central,
but not exclusive, feature. The strategy involves a comprehensive
re-engineering of the entire buy/sell process and the relationships
between government program management and the supply chain.
In addition to using COTS products, the strategy includes·
aggregating demand across organizational and program boundaries
that are traditionally uncoordinated, · determining fundamental
reusable components that may not be recognized as similar in
today's organizational framework, · influencing the design
and creation of products available from the supply chain ·
revamping the mechanisms for matching the buyers and sellers,
i.e., marketplace modernization. Each of these features for
a COTS acquisition strategy is discussed in detail. Case summaries
are presented which demonstrate that implementing these features
will enhance the advantages of using COTS and remove some of
the impediments that have limited early successes. Specific
recommendations are offered to realize the full potential for
COTS utilization through incorporation with re-engineered processes.
Replaceable Components and the Service
Provider Interface
Robert Seacord, Software Engineering Institute
Presentation
Tuesday, February 5, 1:30 p.m. - 2:15 p.m.
A highly touted property of components and component-based software
engineering is the ability to treat components as fully replaceable
units. Commercially successful component models such as EJB,
COM and JavaBeans have not yet produced a marketplace of replaceable
components, while Sun's service provider interface (SPI) has
produced replaceable components in several technology areas.
This paper considers both the meaning of and motivation for
replaceable components, and evaluates the properties of commercially
successful component models and the SPI approach that affects
their ability to support replaceable components.
Rethinking Process Guidance for Selecting
Software Components
Neil Maiden, City University London's School of Informatics
Presentation
Tuesday, February 5, 2:20 p.m. - 3:05 p.m.
This paper reports the results of ongoing research into component-based
software engineering (CBSE) in the European banking sector as
part of the EU-funded BANKSEC project. The importance of complex
non-functional requirements such as dependability and security
presents new generates challenges for CBSE. The paper presents
BANKSEC's vision of an integrated software tool that will provide
process advice for component procurement teams who are tackling
these new problems. The basis for this process guidance is a
situation meta-model that enables the software tool to infer
properties about the current status of the selection process
and recommend process guidance relevant to this situation. This
paper presents the situation meta-model, then demonstrates use
of the model to guide the selection of software components for
banking applications.
Risk Reduction in COTS Software Selection
with BASIS
Keith Ballurio, Software Productivity Consortium
Presentation
Wednesday, February 6, 3:35 p.m. - 4:20 p.m.
Organizations are moving toward COTS-based software development
with limited success. The quality of resulting systems is not
measuring up to expectations. Transitioning to a new development
paradigm requires many kinds of changes, but the most important
concerns the development process. A defined approach tailored
for COTS-integration is needed. The Base Application Software
Integration System (BASIS) is an approach that will improve
the architecture of COTS-based systems by determining the best
integration sequence for the chosen COTS products. The best
sequence is determined by synthesizing objective product evaluations,
emerging practices in integration technologies, and business
priorities. If organizations would tailor their development
process to include BASIS techniques, their transition to COTS-based
development would be much more successful.
The Standard Autonomous File Server: A
Customized, Off-the-Shelf Success Story
Susan Semancik, NASA and Annette Conger, Computer Sciences Corporation
Presentation
Tuesday, February 5, 1:30 p.m. - 2:15 p.m.
The Standard Autonomous File Server (SAFS), which includes both
off-the-shelf hardware and software, uses an improved automated
file transfer process to provide a quicker, more reliable, prioritized
file distribution for customers of near real-time data without
interfering with the assets involved in the acquisition and
processing of the data. It operates as a stand-alone solution,
monitoring itself and providing an automated fail-over process
to enhance reliability. This paper will describe the unique
problems and lessons learned both during the COTS selection
and integration into SAFS, and the system's first year of operation
in support of NASA's satellite ground network. COTS was the
key factor in allowing the two-person development team to deploy
systems in less than a year, meeting the required launch schedule.
The SAFS system has been so successful that it was nominated
for NASA's Software of the Year Award in 1999 and is becoming
a NASA standard resource.
Storyboard Process to Assist in Requirements
Verification and Adaption to Capabilities Inherent in COTS
Sallie Gregor, Joe Hutson, and Colleen Oresky, SAIC
Presentation
Wednesday, February 6, 2:20 p.m. - 3:05 p.m.
One of the challenges of using COTS is defining which requirements
and business processes can be supported with COTS products while
limiting the amount of customization and integration code that
is required. Limiting customization and integration code reduces
risks, lowers maintenance costs, improves ability to upgrade
and the customer more for their investment. To help customers
better understand their requirements and reduce COTS customization,
SAIC successfully used storyboards during the requirements phase.
The storyboard process integrates "use cases" and
screen captures to assist customers in verifying and adapting
their requirements to the capabilities inherent in the COTS
products. An additional bonus of this process is that the screen
captures also support the design phase of the user interface
features. The storyboard process described in this paper was
used to develop large COTS-based systems for customers who historically
developed large custom software applications.
Streamlining the Acquisition Process for
Large-Scale COTS Middleware Components
Anna Liu, CSIRO
Presentation
Wednesday, February 6, 1:30 p.m. - 2:15 p.m.
CSIRO's i-MATE process is an established approach to help IT
organizations in the acquisition of large-scale COTS middleware
components. It aims to minimize technical risk by matching detailed
application and infrastructure requirements to the capabilities
of COTS middleware products. This paper describes a case study
on the use of i-MATE in a project that required the selection
of appropriate components in a compressed timeline and from
a broad range of candidate COTS technologies. The steps and
tools in i-MATE are briefly explained, along with the characteristics
of COTS middleware components that make them a unique challenge
in terms of acquisition and adoption. The case study project
is then outlined, and the key business and technical issues
explained. Finally, we describe and evaluate the enhancements
made to the i-MATE approach to successfully respond to the challenges
encountered.
|
