Expert in Software Systems Scalability, Performance, Reliability, Positive Train Control, Industrial Control Systems, Data and Telecommunications Systems
Expert ID: 738117 New Jersey, USA
Areas of Research
Applications of Markov chains to the study of deadlock problems. Relationship between performance, scalability, architecture, performance requirements engineering, and performance testing. Positive train control and the European Train Control system (ETCS). Scalability of computer systems, including web-based systems and concurrency control methods. Architectural and performance aspects of network management and operations support systems, telecommunication and data communication networks, replicated databases, and webbased systems. Load balancing problems; queueing theory, simulation, and operating systems. Automated recognition of simulation network representations of systems and translation into computationally cheaper analytic models for use in a decision support system. The teaching of performance within software engineering and computer science courses.
Project Areas and Experience
Senior performance architect at a leading supply chain company. Lead performance engineer for a large industrial data analytics project. Investigation of performance, bandwidth, and security needs for Positive Train Control (PTC) and the European Train Control System (ETCS). Lead performance engineer of a project involving agile processes and service-oriented architecture. Investigation of the stability of a building management system. Performance analysis and testing of a web-based configuration system. Performance analysis of a computer-controlled conveyor system. Application flow analysis, architecture review, load testing, and scalability evaluation of a large web-based online banking system. Management and leadership of projects to study the performance and reliability engineering of client/server systems in the customer care and provisioning environments. Scalability and performance evaluation of network management systems, provisioning processes, replicated databases, call centers, and other systems under normal operations and under failure scenarios. Assessment of provisioning processes and systems, particularly as they impact business continuity. Economic studies of call center and call routing design options, with emphasis on reliability and performance under failure. Project leadership of a major architectural exercise for a service-oriented operations
Over 40 technical publications, and nine US patents.
Application of OR techniques to business processes, engineering processes, and the performance evaluation and improvement of computer-driven systems.
• Software performance engineering processes
• Workload analytics and data analysis
• Performance requirements analysis and testing
• Research, development, and patents
• Scalability analysis for software and processes
• Performance troubleshooting
• Simulation and mathematical modeling
• Certified scrum master
|Year: 1984||Degree: PhD||Subject: Computer Science||Institution: Purdue University|
|Year: 1981||Degree: MS||Subject: Computer Science||Institution: Purdue University|
|Year: 1977||Degree: MSc||Subject: Statistics||Institution: University College London|
|Year: 1976||Degree: BSc||Subject: Mathematics||Institution: University of Exeter|
|Years: 2019 to Present||Employer: Undisclosed||Title: Adjunct Professor||Department: School of Systems and Enterprises|
Responsibilities:Instructor for courses in foundations of software engineering, software testing, and software metrics.
|Years: 2017 to 2018||Employer: Ferguson Enterprises (as contractor)||Title: Senior Performance Architect||Department: QA|
Responsibilities:Performance engineering for IT and E-commerce systems in virtualized environments. Led team diagnoses of performance issues, slashing response times. Dramatically reduced the development cost of web performance testing scripts by conceiving and supervising the implementation of a flexible programming method using analytics on traffic data. Mentored and trained junior performance engineers. Established performance governance processes, including architecture, requirements engineering, workload-based performance testing, and the use of performance specifications in procurement. Delivered analytics-based tools for performance management from conception to customer enablement
|Years: 2016 to 2016||Employer: University of L'Aquila, Italy||Title: Visiting Professor of Computer Science||Department: DISIM|
Responsibilities:Lectured on practical aspects of software performance engineering and published research.
|Years: 2003 to 2015||Employer: Siemens Corporate Technology/Corporate Research||Title: Senior Research Scientist||Department: Software Engineering|
Responsibilities:Initiated a US-based consulting practice and research effort in software performance. Developed and applied methodologies for integrating performance engineering methods in the software development process in standard and agile environments. Provided consulting services in software performance, including requirements, architecture, design, testing, and scalability across diverse business units. Negotiated and implemented statements of work, milestones, deliverables, and budgets for engagements. Completed international engagements in Germany, Switzerland, Finland, and India. Obtained 7 patents.
• Lead performance engineer for the Industrial Data Analytics project. Developed and documented workload estimates and performance requirements for digital transformation of predictive maintenance and other applications in industrial and IOT systems. Provided architectural guidance on cloud platform design and measurement. This work included a 3-month rotation in Germany.
• Analyzed data on the European Train Control System to estimate bandwidth requirements for Positive Train Control (jointly work with George Mason University), under a contract with the Federal Railroad Administration.
• Trained a performance testing team in India. Under my guidance, within two weeks, this team identified a show-stopping performance problem in the first load test of a system, and recommended a successful solution.
• Created and taught a pioneering training course on software performance engineering aimed at architects, developers, testers, product managers, prospective performance engineers, and other stakeholders.
• Lead performance engineer on large project to develop a platform based on a service-oriented architecture. Identified and eliminated performance bottlenecks. Provided architectural guidance on early technology choices, thus preventing the use of inappropriate implementation technologies.
• Invented and patented methods to automate the analysis of system performance measurements. Oversaw their implementation.
|Years: 2002 to 2003||Employer: Leapstone Systems, Inc.||Title: Director, System Performance||Department: Office of the CTO|
Responsibilities:Responsible for all performance concerns relating to the Communications Convergence Engine. Oversaw performance testing and analyzed results to assure management and the end customer that performance needs could be met.
• Provided acceptance assurance based on performance tests for the product’s first customer.
• Conceived and developed a configuration sizing tool for marketing staff to make cost-effective recommendations to customers regarding the configuration they should procure.
• Conceived and built a model that predicted that the system would meet reliability requirements.
• Developed performance models to address developers’ specific design issues and make recommendations regarding their implementation.
• Reviewed architecture to provide reassurance that a scaled-down performance testing lab would provide representative performance test results at considerably lower cost than with a production-scale configuration.
|Years: 2001 to 2001||Employer: Hudson Williams, Inc.||Title: Consulting Director||Department: Performance Consulting|
Responsibilities:Senior staff member responsible for supervising clients’ performance tests and for analyzing and interpreting the results.
• Provided guidance on the structure and workloads to be used in clients’ performance tests of web-based financial systems. Analyzed and briefed clients on performance test results.
|Years: 1987 to 2000||Employer: AT&T Bell Labs and successor AT&T Labs||Title: Principal Technical Staff Member||Department: Network Design and Performance Analysis (last of several)|
Responsibilities:Provided performance consulting and research on performance issues in data communications, voice telephony and call center support. Work frequently entailed elicitation of busines processes so that performance needs could be identified and bottlenecks resolved. Obtained 2 U.S. patents.
• Coordinated a multidisciplinary team to identify ways to digitize service and equipment provisioning processes intended to replace diverse and conflicting legacy systems and manual processes across multiple business units.
• Designed a test plan and carried out a performance study of a Citrix-based thin client NT server system and a Cold Fusion-based system with an Oracle back end to determine current capacity and
the need for future system growth. Revealed the presence of software bottlenecks that prevented adequate utilization of the existing hardware platform. Procurement of additional hardware was not recommended until the bottlenecks could be fixed. This delay resulted in savings in hardware cost.
• Developed a sizing tool for AT&T’s implementation of HP OpenView. Invented a patented method to remedy a blocking problem that caused HP OpenView to freeze when a network failure made multiple nodes unreachable.
• Specified functional requirements for a call routing feature, and provided technical support from development to explaining the mechanism to the operations support personnel who installed feature.
|Years: 1984 to 1987||Employer: University of California, Santa Barbara||Title: Assistant Professor||Department: Computer Science|
Responsibilities:Conducted research on queueing problems in computer communication networks. Advised students.
• Developed and taught courses in operating systems principles, simulation and modeling, and computer performance evaluation.
• Taught an undergraduate course in computer architecture.
• Supervised a master’s student who successfully completed his dissertation in less than one year.
• Published papers in conferences and journals.
|Years: 1977 to 1979||Employer: Logica Ltd.||Title: Consultant||Department: Data Processing Economics|
Responsibilities:Consultant on a specialist team engaged in the performance evaluation of computer systems. Work included hardware monitoring, the use of the BEST/1 ™ queueing network performance analysis package, and advising colleagues on the suitability of modeling techniques.
|Years: 1986 to 1986||Employer: National Bureau of Standards and Technology (now NIST)||Title: Visiting Scientist||Department: Systems and Network Architecture Division|
Responsibilities:Visiting scholar. Conducted research on performance models of front-end protocol processors which was subsequently published. Taught a basic course in computer performance modeling.
|Years: 1983 to 1984||Employer: Pritsker and Associates||Title: Consultant||Department: R&D|
Responsibilities:Conceived and implemented an automatic translator to convert network-oriented system representations for discrete event simulation into analytical queueing models, which are computationally less expensive. The tool was developed using parsing techniques that are often used in compilers. This work was part of the US Air Force IDEF2 decision support system, and was subsequently published.
|Years: 1982 to 1982||Employer: IBM Thomas J. Watson Research Center, Yorktown Heights||Title: Research Assistant||Department: Protocol Network Analysis|
Responsibilities:Analyzed the performance of networking configurations using the RESQ™ simulation package.
|Years: 1981 to 1981||Employer: BGS Systems, Inc.||Title: Research Assistant||Department: R&D|
Responsibilities:Conducted research on priority queues which was subsequently published. Visit was under the auspices of Purdue University and sponsored by the National Science Foundation.
|Years: 1979 to 1984||Employer: Purdue University||Title: Teaching Assistant, Research Assistant||Department: Computer Science|
Responsibilities:Teaching assistant for a graduate level course on operating systems. Research on priority queues and operational analysis under an NSF grant.
Research on human factors; provided statistical analysis. Taught an introductory course on FORTRAN programming.
|Years||Country / Region||Summary|
|Years: 1971 to 1979||Country / Region: England||Summary: Two years' consulting experience in London, two university degrees, last two years of secondary school. Member of local chamber orchestra while living in London.
Consulting experience: member of a specialist team measuring and modeling commercial computing systems based on mainframes. Provided UK tech support for the BEST/1 queueing network analysis package.
|Years: 2010 to 2011||Country / Region: Switzerland||Summary: Assigned to provide performance consulting to Siemens business unit. Also spent school and university vacations in Switzerland when parents were living there, so know quite about local customs and even speak a little of the local dialect.|
|Years: 2005 to 2006||Country / Region: Germany||Summary: One year assignment at Siemens business unit in Munich. Recalled to visit the business unit on multiple occasions subsequently. Multiple visits to Germany on behalf of Siemens 2004-2014, including a 3-month assignment in 2014 on the Industrial Data Analytics Project, and multiple visits to a business unit in Erlangen in 2010. Socially conversant in German. Sang in a chorus for an entire season in Munich.|
|Years: 2007 to 2007||Country / Region: India||Summary: Three-week assignment to a Siemens business unit. Trained a local performance testing team which Expert organized using a directed form of Scrum.|
|Associations / Societies|
|Fellow of the Royal Statistical Society, 1977-present.
Member of Association of Computing Machinery 1981-present.
Treasurer of the Red Bank Chamber Music Society
|Licenses / Certifications|
|Certified Scrum Master, 2008.|
|Steering Committee Co-Chair, International Conference on Performance Engineering
Steering Committee member
(undisclosed position), International Conference on Performance Engineering (then called WOSP/SIPEW)
Organizing committees and reviewer for several international conferences.
|Awards / Recognition|
|• A. A. Michelson Award. Given by the Computer Measurement Group
• Association of Computing Machinery Recognition of Service Award for contributions to ACM as Co-Chair WOSP/SIPEW’10: First Joint WOSP/SIPEW Conference on Performance Engineering, San Jose, California
• Co-Recipient (with Alberto Avritzer and Elaine Weyuker), ACM SIGSoft Distinguished Paper Award, ACM Workshop on Software and Performance
|Publications and Patents Summary|
|9 US patents relating to computer performance, network management, software rejuvenation, and call center management.
Over 40 publication on computer networking, priority scheduling, queueing models and simulations of computer systems, replicated databases.
Invited Talks: at conferences: 8; other external presentations: 15; talks at Siemens Workshops: 6
|Expert Witness Experience|
|Deposed in complicated art restitution case, defendant in rem, and as plaintiff in a personal injury case.|
|Training / Seminars|
|Developed a corporate training course on software performance engineering.
Have taught the following courses at the university level:
• At Employer: Foundations of Software Engineering, Software Testing, Software Metrics.
• At the University of L’Aquila: Guest lectures on performance testing and measurement as part of a course on performance modeling and software quality assurance.
• Siemens Corporate Technologies: developed a short course on software performance engineering. The course was presented at Siemens installations in Germany, India, and Finland.
• At UC Santa Barbara: Operating Systems Principles, Foundations of Computer Architecture, Discrete Event Simulation and Modeling, Computer Systems Performance Evaluation and Modeling.
• At Purdue University: Teaching assistant for a graduate level operating systems course and for a first course in FORTRAN programming for engineers.
At AT&T Labs, led a team to prepare a document on performance aspects of vendor management. This contained guidelines on specifying performance requirements and evaluating the architectures of systems to be procured from outside vendors.
|At Leapstone Systems, developed a configuration sizing tool to assist the marketing department with the preparation of cost-effective system configurations to meet customer traffic requirements.|