PROGRAMME

Tuesday, 3 November 2015

Session1

Session2

Session3

Abstracts

Evaluating open systems dependability in mission-critical domains

Alberto Avritzer

A mission-critical system is understood to be a system that is central to a specific domain and might impact business continuity. Specifically, we focus on events causing significant disruption or multiple failures to the domain. We present three examples of survivability analysis in different domains : 1) smart-grid, 2) distributed software development, and 3) data streaming. The smart-grid domain is illustrated by disruptions caused by large storms, as for example the Sandy storm impact on the NY metropolitan area. In the distributed software development domain, loss of communication and trust between distributed teams can be seen as a major disruption to the software development business. In the data streaming domain, failures of software architecture components can cause significant disruption to the data streaming operations. In all cases, we present the metric, the modeling approach, and the benefits of using the proposed modeling approach for the assessment of the alternative investments in system and software architecture, as assessed by the defined survivability.

Alberto Avritzer received a Ph.D. in Computer Science from the University of California, Los Angeles, an M.Sc. in Computer Science for the Federal University of Minas Gerais, Brazil, and the B.Sc. in Computer Engineering from the Technion, Israel Institute of Technology. He is a consultant in software reliability and performance assessment of mission critical systems. He was a Senior Member of the Technical Staff in the Software Engineering Department at Siemens Corporate Research, Princeton, New Jersey for 11 years, where he worked on the assesment and improvement of software reliability and software performance of mission critical systems in the healthcare, transportation and building technology domains. Before moving to Siemens Corporate Research, he spent 13 years at AT&T Bell Laboratories, where he developed tools and techniques for performance testing and analysis. He spent the summer of 1987 at IBM Research, at Yorktown Heights. His research interests are in software engineering, particularly software architecture and testing, monitoring and rejuvenation of smoothly degrading (aging) systems, and metrics to assess software architecture, and he has published over 50 papers in journals and refereed conference proceedings in those areas. He is a Senior Member of ACM.

The challenge of assuring autonomous systems

Robin Bloomfield (City University London, Adelard):

Open systems dependability standardization activity in IEC TC56 dependability

Yoshiki Kinoshita (Kanagawa):

I will give an overview of the development of IEC 62853 Open Systems Dependability and activities on relevant standard such as SO/IEC/IEEE 15288 System Life Cycle Processes and ISO/IEC 15026 Systems and Software Assurance (Part 1 - 4).

A case study of enterprise application development using DEOS Forwarding Process Tools

Tatsumi Nagayama (Symphony)

The impact of “openness" on safety critical FPGA based system

K Netkachova (Adelard and City Univeristy London), Vyacheslav Kharchenko (KhAI) and Vladimir Sklyar (Radiy):

Ecosystem-based assurance for plug-and-play medical systems

Oleg Sokolsky (Pennsylvalnia)

A Petri net model of the DEOS life cycle

Makoto Takeyama (Kanagawa)

The talk discusses a formal life cycle model of the DEOS Process. The model serves as a context for formal assurance arguments that instances of the DEOS Process achieves open systems dependability. The DEOS Process is typically explained using its characteristic double-loop life-cycle figure consisting of Failure Response Cycle and Change Accommodation Cycle. However, the sense of the double loop is left informal and needs to be formally modelled to support OSD assurance argument rigorously. The talk proposes a model that represents the double loop as a kind of Petri net (predicate/transition net) to be used as an organising framework for describing requirements for life cycle processes, evidence of satisfaction and arguments for OSD achievement.