Abstract: Emphasis on system safety and reliability assessment allowed us to place societal trust on mass-transportation and medical systems, energy, space, financial and military infrastructures, and many other critical domains.  As technology intrudes into all aspects of our lives, existing verification and validation approaches are not sufficient anymore.  In addition to meeting safety, reliability, and security standards, we expect systems to protect our privacy, uphold equity, integrity and human rights, perform without biases, and meet other stated and unstated expectations.  Further, some of the logic embedded in software may be the product of machine learning, making it unintuitive for humans to understand the rules it encodes, thereby difficult to assess.
In this talk, we will discuss software accountability principles.  Efforts to regulate software and services are increasing, with the goal of making organizations and individuals more accountable for its consequences.  But what constitutes accountable software is not understood well.  We will overview the outcomes of traditional software safety assessment processes and compare them with emerging verification needs.  Examples from adaptive flight controls and airport passenger management will offer preliminary illustrations of the dimensions of software accountability assurance. Our talk will conclude by outlining opportunities for future research.

Bio: Bojan Cukic is a Professor and Interim Dean in the College of Computing and Informatics at the University of North Carolina at Charlotte. He previously served as Associate Dean for Academic Programs and Student Success as well as the Chair of the Department of Computer Science and the Interim Executive Director of the UNC Charlotte Data Science Initiative. Dr. Cukic’s research interests include information assurance and biometrics, software engineering with emphasis on verification and validation, and resilient computing. He received MS and PhD degrees in Computer Science from the University of Houston, and holds an honorary doctorate from the University of Rijeka, Croatia.

Abstract:  Non-Markovian models enable the representation of complex cyber-physical systems including synchronous phenomena (e.g., periodic arrivals, timeouts, offsets) and non-exponential durations possibly with bounded support (e.g., delays in transportation systems, minimum and maximum execution times in real-time systems), notably facilitating fitting of stochastic parameters from observed data. Solution techniques for quantitative evaluation of these models can give decisive support to system development by enabling early verification of design choices and assessment of non-functional requirements. Effective exploitation of this potential faces both practical and theoretical complexities. On the one hand, the activity of model construction must fit the context of use, automatically generating models from artifacts of the industrial practice and from actual operational data. On the other hand, the analysis needs to be able to afford complexity in terms of the size and class of the stochastic processes underlying the models generated automatically without manual intervention.
This talk presents a Model Driven Engineering (MDE) approach to performance evaluation of multimodal urban intersections, supporting automated transformation of semi-formal artifacts of the Unified Modeling Language (UML) into stochastic models that can be efficiently analyzed. On the one hand, a meta-model of the system captures structural and behavioral information on the tram flows and the car flows that cross at the intersection. On the other hand, transient analysis of a microscopic model of tram traffic, defined by Stochastic Time Petri Nets (STPNs), is combined with analytical solution of a set of ordinary differential equations characterizing the behaviour of a macroscopic model of car traffic, defined by finite-capacity queues, providing the expected number of queued cars over time for intervals of arbitrary duration. Validation of the approach in terms of accuracy and complexity is performed with respect to a traffic simulator using a large-size benchmark. Effectiveness of the solution is demonstrated on relevant use cases in operation and management of urban transportation systems.

Bio:  Carnevali Laura is associate professor of computer science at the Department of Information Engineering of the University of Florence, where she is a member of the Software Technologies Lab. Her research is focused on formal methods for model driven development of real-time software and on solution techniques for performance evaluation of concurrent non-Markovian models. As a part of her research, she has contributed to the implementation of the ORIS Tool.

Abstract:  TBD

Bio:  Domenico Cotroneo is currently full Professor at Department of Electrical Engineering and Information Technology (DIETI, www.dieti.unina.it). He is IEEE senior member and elected member of IFIP WG 10.4, Dependable Computing and Fault Tolerance. His research interests include software fault injection, dependability assessment, and field-based measurement techniques.

Abstract:  TBD

Bio:  Enrico Vicario is a Full Professor of Computer Science and Engineering. Since November 2016, he is the Head of the Department of Information Engineering of the University of Florence. He works in the area of Software Engineering, at the at the Software Technologies Lab of the University of Florence, with a present scientific focus on: 1) model based development, verification, and evaluation of concurrent systems with uncertain temporal parameters and stochastic durations. 2) software architectures and software engineering methods.

Abstract:  Talking about our efforts in mobile application offloading, where many decisions have to be taken and several concerns have to be met. This is reflected in the metrics that are considered and they can be combined in different ways, leading to different outcomes.

Bio:  Katinka Wolter currently works at the Department of Mathematics and Computer Science, Freie Universität Berlin. Katinka does research in Computer Security and Reliability, Computer Communications (Networks) and Distributed Computing. One of her group's current projects is 'Reliability and scalability of Streaming'. With Will Knottenbelt from Imperial College London she is working on fairness of cryptocurrency mining pools.

Abstract:  The talk describes model checking based on algebraic specifications of systems/protocols. Maude is used as a formal specification language. Maude is equipped with model checking facilities: a reachability analyzer (the search command) and an LTL model checker, where LTL stands for linear temporal logic. Two flawed versions and one correct version of a mutual exclusion protocol are used as examples. The talk mentions one of the most annoying problems in model checking: the state space explosion problem. The talk also mentions a possible approach to alleviating the problem.

Bio:  Kazuhiro Ogata received the B.S., M.S., and Ph.D. degrees in engineering from Keio University, in 1990, 1992, and 1995, respectively. He is currently a Professor with the Japan Advanced Institute of Science and Technology (JAIST) and the director of Research Center for Advanced Computing Infrastructure in JAIST. His research interests include applications of formal methods to systems/protocols, such as post-quantum cryptographic protocols and quantum systems/protocols.

Abstract:  Complex systems in different domains contain significant amount of software. Several studies have established that a large fraction of system outages are due to software faults. Traditional methods of fault avoidance, fault removal based on extensive testing/debugging, and fault tolerance based on design/data diversity are found inadequate to ensure high software dependability. The key challenge then is how to provide highly dependable software. We discuss a viewpoint of fault tolerance of software-based systems to ensure high dependability. We classify software faults into Bohrbugs and Mandelbugs, and identify aging-related bugs as a subtype of the latter. Traditional methods have been designed to deal with Bohrbugs. The key challenge then is to develop mitigation methods for Mandelbugs in general and aging-related bugs in particular. We submit that mitigation methods for Mandelbugs utilize environmental diversity. Retry operation, restart application, failover to an identical replica (hot, warm or cold) and reboot the OS are reactive recovery techniques applied after the occurrence of a failure. They are examples of techniques that rely on environmental diversity. For software aging related bugs, it is also possible to utilize a proactive environmental diversity technique known as software rejuvenation. We discuss environmental diversity both from experimental and analytic points of view and cite examples of real systems employing these techniques.

Bio:  Kishor S. Trivedi holds the Hudson Chair in the Department of Electrical and Computer Engineering at Duke University, Durham, NC. He has a B.Tech. (EE, 1968) from IIT Mumbai, M.S. (CS, 1972) and PhD (CS, 1974) from the University of Illinois, Urbana-Champaign. He has been on the Duke faculty since 1975. He is the author of a well-known text entitled, Probability and Statistics with Reliability, Queuing and Computer Science Applications, first published by Prentice-Hall; a thoroughly revised second edition of this book has been published by John Wiley. He has authored several other books, the most recent being Reliability and Availability Engineering by Cambridge University Press in 2017. He is a Life Fellow of the Institute of Electrical and Electronics Engineers. He is a Golden Core Member of IEEE Computer Society. He has published over 600 articles and has supervised 48 Ph.D. dissertations. His h-index is 110. He is a recipient of IEEE Computer Society Technical Achievement Award for his research on Software Aging and Rejuvenation. He is a recipient of IEEE Reliability Society’s Lifetime Achievement Award.

Abstract:  Cloud-assisted body area networks have been the focus of researchers in past years as a response to the development of robust wireless body area networks (WBANs). However, existing works mostly focus on WBAN applications and data analysis. In this talk, we will introduce a cloud-centered heterogeneous and comprehensive wireless body sensor data collection, streaming, and analytics framework. The system combines the sensor control and data aggregator event detection, real-time data analysis, visualization, and streaming into one Android App and incorporates four key components in the cloud server: data repository, algorithm repository, machine learning engine, and web portal. The framework also includes a transport layer progressive data transmission algorithm as well as a dynamic EAP based IEEE 802.15.6 MAC protocol.

Bio: Ming Li is currently a Professor in the Department of Computer Science, California State University, Fresno. He served as the department chair from August 2013 to August 2017. Prior to that, he was an Assistant and then Associate Professor from August 2006 to 2018. He received his M.S. and Ph.D. degrees in Computer Science from The University of Texas at Dallas in 2001 and 2006, respectively. His research interests include QoS strategies for IEEE 802.11 wireless LANs, mobile ad-hoc networks, and heterogeneous wired and wireless networks, multimedia streaming over wireless networks, body area networks, and robot swarm communications.

Abstract: Artificial intelligence (AI) applications in industry have been growing rapidly in the past few years. Many traditional systems have become more intelligent through the development of internet technologies, together with new artificial intelligence and machine learning approaches. However, there are new and challenging problems, especially with the software aspects of AI applications. In fact, how to measure the reliability of AI is unclear to begin with, not to say how to develop reliable AI technology and systems. In this talk, we will share some thoughts and experiences from a systems engineering perspective and discuss some of these issues. More research on software and hardware reliability still needs to be carried out. Data accuracy and sensor reliability are some of the related issues that should be paid attention to, so that software reliability can be predicted more accurately over time. Testing of complex systems, in the context of system upgrading or updating, should be planned carefully as well.

Bio:  Prof. Min Xie completed his undergraduate study and received his MSc from the Royal Institute of Technology, Stockholm, and his PhD in 1987 was from Linkoping University, Sweden. Dr Xie joined the National University of Singapore in 1991 as one of the first recipients of the prestigious Lee Kuan Yew Research Fellowship. After 20 years there, he moved to City University of Hong Kong as Chair Prof of Industrial Engineering in 2011. Prof Xie serves as an editor, associate editor and on the editorial board of 20 established international journals. Prof Xie has supervised more than 60 PhD students and they hold regular positions in banking, industry and academia in different continents. Prof Xie has published over 300 journal papers and 10 books, and he was elected fellow of IEEE in 2005 for his contribution to software and systems reliability. He was also elected Academician to European Academy of Sciences and Arts earlier this year.

Abstract:  Software reliability is intrinsically a quantitative discipline. However, we know from practice that qualitative factors play an important role in this field. Since 2009, Dr. Rivalino Matias has been investigating the concept of QoE applied to software reliability engineering. As a result, a body of knowledge has been built from both conceptual and practical perspectives. In this talk, Dr. Matias will share research findings on the use of QoE in software reliability, highlighting some pitfalls and biases that are observed in practice. Case studies involving real-world software systems will be discussed.

Bio:  Rivalino Matias, Jr. is currently a tenured associate professor in the Computing School at the Federal University of Uberlandia, Brazil. In the last twenty and two years, he has been studying different aspects of software reliability. Dr. Matias has served as a member of the steering and organizing committees in several high-rank international conferences, as well as an editor and a reviewer of many prestigious international journals and conferences. His main research interests include the reliability of operating systems and theoretical and practical aspects of software aging and rejuvenation. Dr. Matias also worked for world-wide companies such as Microsoft, IBM, and Red Hat.

Abstract:  With the rise of software complexity, software-related accidents represent a significant threat for computer-based systems. Software Fault Injection is a method to anticipate worst-case scenarios caused by faulty software, through the deliberate injection of software faults (“bugs”). This presentation will present an overview on Software Fault Injection, and will discuss how it evolved to achieve fault representativeness, efficiency, and usability. Finally, this presentation will introduce recent developments on customizing and analyzing fault injection experiments with machine learning.

Bio:  Roberto Natella is assistant professor at the Federico II University of Naples, Italy, and co-founder of the Critiware s.r.l. academic spin-off company. His research interests include software fault injection, security/robustness testing, dependability benchmarking, and software aging and rejuvenation, and their applications in operating systems and virtualization technologies. His work has been supported by national, European, and industry-funded research projects in cooperation with Leonardo-Finmeccanica, CRITICAL Software, and Huawei Technologies. He authored more than 60 publications in journals and conferences on dependable computing and software engineering. He has been on the steering committee of the IEEE International Workshop on Software Certification (WoSoCer), and Program Committee Chair of the IEEE International Symposium on Software Reliability Engineering (ISSRE).

Abstract:  The talk will review the interplay between software testing and reliability. It will give an overview of testing strategies for both the problem of reliability assessment and of reliability improvement. The reliability-driven planning of testing effort is also discussed for modular software systems. Finally, some advanced techniques for making reliability testing more efficient will be outlined.

Bio:  Roberto Pietrantuono received the PhD degree in computer and automation engineering from the Federico II University of Naples, Italy, in 2009. He is assistant professor with the Federico II University of Naples, Italy. His research interests include the area of software reliability engineering, particularly in the software verification of critical systems, software testing, and software reliability analysis. He is a senior member of the IEEE.

Abstract:  In this talk I introduce the recent development of software reliability models proposed in our research group. The non-homogeneous Poisson process (NHPP)-based software reliability models (SRMs) are the most classical stochastic models to describe the software fault count process in the system testing phase. Over the last four decades, a great number of NHPP-based SRMs have been developed in the literature, where almost all of them are based on the representative software-fault-detection time distributions. More specifically, these distributions belong to the well-known Pearson type or extreme-value distributions, which are quite familiar to the lifetime analysis. In the first part of my talk, I introduce another distribution family called the Burr type distributions. In a comprehensive numerical experiment, I show that our Burr-type NHPP-based SRMs could provide better goodness-of-fit and predictive performances than the existing NHPP-based SRMs. In the second part, I generalize the NHPP-based SRMs to the non-homogeneous Markov processes (NHMPs), and propose two extended NHMP-based modeling frameworks; generalized binomial process (GBP) and generalized Polya process (GPP), to describe the software fault count processes with general software-fault-detection time distributions. I carry out the comparative study between NHPP-based SRMs and NHMP-based SRMs through numerical illustrations, and point out that beyond NHPPs should be a useful modeling paradigm.

Bio:  Since 2002, Tadashi Dohi (Member, IEEE) has been a Professor with Hiroshima University, where he has been appointed as the Vice Dean of the School of Informatics and Data Science, since 2018. His research interests include reliability engineering, software reliability, and dependable computing.,Dr. Dohi is a member of the Operations Research Society of Japan, the Institute of Electrical, Information and Communication Engineers, the Information Processing Society of Japan, and the Reliability Engineering Association of Japan. He is also a member of the Editorial Board of the IEEE Transactions on Reliability.

Abstract:  In this seminar, Dr. Huang will talk about several research projects of my group to make parallel and heterogeneous computing easier to handle, with a specific focus on CAD applications. Dr. Huang will present a novel programming environment and a learning-based runtime that streamline the implementation complexity of large heterogeneous programs. For example, they have shown up to 500x speed-up using their system to accelerate a circuit simulation workload. Their software releases have accumulated over 1.5M downloads, being used by several academic and industrial projects.

Bio:  Dr. Huang is an assistant professor in the ECE Department at the University of Utah. He received his PhD from the ECE Department at the University of Illinois at Urbana-Champaign and BS/MS from the CS Department at Taiwan’s National Cheng Kung University. His research group has been creating software systems to streamline the building of high-performance computing applications, including machine learning, computer-aided design, and quantum computing.