SaS Seminars 2015
Software and Systems Research Seminar Series
The SaS Seminars are a permanent series of open seminars of the Division of Software and Systems (SaS) at the Department of Computer and Information Science (IDA), Linköping University. The objective of the seminars is to present outstanding research and ideas/problems relevant for SaS present and future activities. In particular, seminars cover the SaS research areas software engineering, programming environments, system software, embedded SW/HW systems, computer systems engineering, realtime systems, parallel and distributed computing, and theoretical computer science. - Two kinds of seminars are planned:
talks by invited speakers not affiliated with SaS,
internal seminars presenting lab research to whole SaS.
The speakers are expected to give a broad perspective of the presented research, adressing the audience with a general computer science background but possibly with no specific knowledge in the domain of the presented research. The normal length of a presentation is 60 minutes, including discussion.
The SaS seminars are coordinated by Christoph Kessler.
Recent / Upcoming SaS Seminars (2015)
A class of precomputation-based distance-bounding protocols
Prof. Sjouke Mauw, University of Luxembourg
Friday, 18 Dec. 2015, 13.15, room Donald Knuth
Distance-bounding protocols serve to thwart various types of proximity-based attacks, such as relay attacks. A particular class of distance-bounding protocols measures round trip times of a series of one-bit challenge-response cycles, during which the proving party must have minimal computational overhead. This can be achieved by precomputing the responses to the various possible challenges. In this paper we study this class of precomputation-based distance-bounding protocols. By designing an abstract model for these protocols, we can study their generic properties, such as security lower bounds in relation to space complexity. Further, we develop a novel family of protocols in this class that resists well to mafia fraud attacks.
Side-Channel Analysis for Embedded Devices: Challenges and Pitfalls
Associate Prof. Lejla Batina, Radboud University Nijmegen, The Netherlands
Tuesday, 27 October 2015, 15:15, room John von Neumann
In this talk we give an overview of side-channel attacks and some well-known countermeasures. As an example, we discuss S-boxes of block-ciphers that are usual targets of side-channel attacks and it is an interesting challenge to come up with design techniques to improve their side-channel resistance. Several metrics were proposed in attempt to characterize the resilience of S-boxes against DPA, most notably the transparency order metric. Recently, some new metrics and new results appear showing that this problem is far from being solved.
Lejla Batina is an Associate Professor at Radboud University Nijmegen, The Netherlands and a free researcher at KU Leuven, Belgium. She received her M.Sc. degree in Mathematics from the University of Zagreb, Croatia in 1995 and Ph.D. degree in engineering from the K.U. Leuven in 2005. She also worked for three years as a cryptographer for SafeNet B.V. in The Netherlands. Her research interests include efficient arithmetic for cryptography, secure implementations of cryptographic algorithms, side-channel security, lightweight cryptography e.g. crypto for RFIDs, sensor networks, etc.
Compositional Predictability Analysis of Mixed Critical Real Time Systems
Dr. Abdeljalil Boudjadar, RTSLAB, Linköping University, Sweden
Monday, 5 October 2015, 10:15, room Alan Turing
Predicatability has been identified as a related input sensitive requirement which ascertains that the externally observable behaviour of a component remains the same despite internal non-determinism while removing external non-determinism (i.e., keeping the inputs to the component and their timing unchanged). Proving the predictability means that the system analysis is successfully passed regarding both data flow and time-constrained behavior under any execution assumption. Different techniques have introduced to analyze the predictability of real-time systems, where the analysis does not leverage system structure and systems are analyzed monolithically. This may lead to a state space explosion, making large systems non-analyzable. In this talk, I will introduce a compositional framework for analyzing the predictability of component based embedded real-time systems and show how to support the predictability through certain design restrictions. The framework utilizes automated analysis of tasks and communication architectures to provide insight on the schedulability and data flow. The system architecture we consider in this paper is structured in terms of components having different criticality levels.
Abdeljalil Boudjadar just started his postdoc with RTSLAB, Linköping University. He obtained his PhD in Computer Science from the Paul Sabatier University in Toulouse-France where he worked on "Compositional Semantics and Refinement of Timed systems". He continued with a postdoc at Aalborg University-Danmark followed by another postdoc at Queen's University-Canada.
Constraint Programming in Code Generation for Custom Computing Platforms
Prof. Krzysztof Kuchcinski, Lund University, Sweden
Thursday, 23 April 2015, 10:30, room Alan Turing
Custom processor architectures are often proposed as more efficient alternatives to general purpose processors in terms of performance and power. They are often multi-core computing platforms with specialized processors offering often data-level parallelism. This makes it difficult to program them, and it is often too expensive to develop specialized compilers. Constraint programming offers a quick way to define architectural constraints together with program constraints in a single model. Solutions to such models deliver executable code.
In this talk, we will introduce quickly basics of constraint programming and an example of a custom architecture. We will then concentrate on mapping and code generation problems and show several possible solutions and results.
The presented results are part of the "HiPEC: High Performance Embedded Computing" project funded by SSF.
Krzysztof Kuchcinski holds a chair in Computer Science at Lund Institute of Technology (LTH), Lund, Sweden. His main research interests include different aspects of digital system analysis and synthesis as well as applications of constraint programming. See his home page at cs.lth.se/krzysztof_kuchcinski
Instituto Tecnologico de Aeronautica - ITA: the challenges of its expansion
Wednesday, 1 April 2015, 13:15, room Alan Turing
Instituto Tecnologico de Aeronautica - ITA is a higher education institute of Brazil whose mission is to conduct research and teaching activities to develop the Brazilian aerospace sector. After 65 years of activities, the most prominent result of ITA is Embraer. Embraer, a Brazilian company, is the third largest commercial aviation manufacturer in the world. ITA is claimed to have helped to accomplish several other Brazil's strategic programs, such as the organization of the telecommunications sector, the Brazilian aviation infrastructure, R&D in the space sector, and the creation of the defense industry. The reasons of the successes of the institute have been attributed to two educational factors: very competitive admission process of students and strict teaching system. However, there is an understanding that those factors are not sufficient. Currently ITA is undertaking an expansion program whose initial goal is to double the number of students in five years horizon. Additionally to the infrastructure development and human resources' hiring, the ITA's expansion is an opportunity to re-think the institute and consider other factors. The expansion is being seen as challenges in three areas: engineering education - adoption of more effective and stimulating models of education in engineering; research - improvement of effectiveness of research; and innovation - improvement of interactions between ITA and strategic sectors to generate innovation. In the talk, I will present and discuss how ITA is facing these challenges.
Resource Contention Management in Virtualized Systems
Diwakar Krishnamurthy, Associate Professor, University of Calgary, Canada
Friday, 13 March 2015, 13:15, room Alan Turing
Remark: This seminar is a joint SaS/ADIT seminar.
Public and private cloud computing environments employ virtualization methods to consolidate application workloads onto shared servers. Modern servers typically have one or more sockets each with one or more computing cores, a multi-level caching hierarchy, a memory subsystem, and an interconnect to the memory of other sockets. While resource management methods may manage application performance by controlling the sharing of processing time and input-output rates, there is generally no management of contention for virtualization kernel resources or for the memory hierarchy and subsystems. Yet such contention can have a significant impact on application performance. Hardware platform specific counters have been used for detecting such contention. In this talk, we show that such counters alone may not be always sufficient for detecting contention. We propose a software probe based approach for detecting and mitigating contention for shared platform resources. Results from a private cloud and a public cloud show that the probe imposes a low overhead and is remarkably effective at detecting and mitigating performance degradations due to resource contention over a wide variety of workload scenarios and platform architectures.
Diwakar Krishnamurthy is currently an associate professor and director of the software engineering program in the department of electrical and computer engineering at the University of Calgary, Calgary, Canada. His research interests are broadly focused on computer systems performance evaluation. In the past, his research team has contributed techniques to test, model, optimize, and manage the performance of enterprise application systems. He is currently involved in projects related to runtime management of cloud systems, simulation techniques for studying healthcare systems, and big data systems. His research has been supported by HP Labs and SAP Research.
Combating Unpredictability in Multicores through the Multi-Resource Server
Dr. Rafia Inam, Mälardalen Univ., Sweden
Wednesday, 25 Feb. 2015, 13:15, room John von Neumann
This seminar presents challenges that hinder the predictable integration and execution of real-time applications on multicore platforms. It presents how shared resources, like CPU, memory-bus bandwidth, caches, and memory cause unpredictability and interference. We propose to adapt the traditional server-based scheduling approach on the multicore platforms with additional resource-reservations to control the shared access to such resources and present the multi-resource server (MRS) as a solution such that the execution of real-time applications becomes predictable.
A multi-resource server uses resource reservation for both CPU bandwidth and memory-bus bandwidth to bound the interferences between tasks running on the same core, as well as, between tasks running on different cores. The latter could, without MRS, interfere with each other due to contention on a shared memory-bus and memory. The results indicate that MRS can be used to "encapsulate" legacy systems and to give them enough resources to fulfill their purpose. Further, the compositional schedulability analysis for MRS is also provided and an experimental study is performed to bring insight on the correlation between the server budgets.
Keywords: Hierarchical scheduling, memory-bus bandwidth, cache coloring, memory-bank partitioning.
Rafia Inam recently obtained a PhD degree from Mälardalen University, Sweden. Her PhD research was on hierarchical scheduling for real-time embedded systems. A main focus of her work is on predictable execution of real-time systems for unicore and multicore platforms.
Network Challenges in Cyber-Physical Systems
Prof. Luis Almeida, U. Porto, Portugal, and Mälardalen Univ., Sweden
Tuesday, 27 Jan. 2015, 13:15, room John von Neumann
Cyber-Physical Systems (CPS) frequently rely on networking infrastructures. These necessarily play a central role in supporting the needed system-wide properties, being timeliness a particularly important one as dictated by the dynamics of the associated physical processes. In this talk, we present the case for flexible channel reservation-based communication as a means to provide scalable, open and adaptive latency-constrained communication and thus enable an efficient design of emerging CPS applications, such as Remote Interactions, Collaborative Robotics and other CPS applications with variable and unanticipated bandwidth requirements.
Luis Almeida is an associate professor at the ECE Dep. of the University of Porto and a member of the Institute of Telecommunications in Porto where he coordinates the Distributed and Real-Time Embedded Systems group, working in real-time communication protocols and middleware for embedded systems and cooperative robotics. Currently he is also guest professor at the School of Innovation, Design and Engineering of Mälardalen University, in the Embedded Systems Division.
Previous SaS Seminars
Page responsible: Christoph Kessler
Last updated: 2016-03-01