Skip to content. Skip to navigation
McGill Home SOCS Home
Personal tools
You are here: Home Announcements and Events Colloquium Archive

Colloquium Home
Fall 2013 Schedule
Winter 2014 Schedule
How to Suggest a Speaker
Archives

 

Date
( Fall 2011 )
Speaker and Abstract
2011/09/16 Speaker: Manfred Nagl
Affiliation: RWTH Aachen University
Area: Software Engineering
Title: Integrators for Consistency Management or How to Keep the Results of Different Developers Consistent?
Abstract: Within the Collaborative Research Center 476 IMPROVE and the Transfer Center 61, both funded by German Research Foundation DFG (comparable to a center of excellence), we have studied novel tool support for design processes in Chemical Engineering. Both projects were carried out between groups of Engineering and Computer Science at RWTH Aachen University. The concepts and results are also applicable to software construction processes. One subproject deals with the question to keep the results of different developers consistent to each other. Consistency or traceability is a current topic, especially in connection to model-driven development. We have been studying interactive and incremental tools for achieving consistency, called integrators, since many years. Tools are systematically derived using a meta tool environment, being based on graph specifications. Rules for consistency management can be postulated before the tool construction process but also by the user during tool application. Our findings are being used in industry. In the first part of the talk we report about the activities in the Software Engineering group. In the second part, we give an overview of the CRC and the TC activities. In the third part, we discuss our findings on integrator tool construction and maintenance. Biography of Speaker:

Prof. Nagl is author of five books and editor/coeditor of about 20 books. He authored/ coauthored 125 articles in scientific journals, proceedings, and books. He authored the first book on graph rewriting, and also on software architectures worldwide, and also the first Ada book Germany-wide. He was speaker of the DFG research group SUKITS (1990-1997), he was speaker of the DFG Collaborative Research Center SFB 476 (1997-2006), and speaker of the Transfer Center TB 61, also financed by DFG, 2006-2009. SUKITS was on tool integration in mechanical engineering, the SFB and the TB was on tool integration in chemical and plastics engineering. The books on these projects are outstanding. At RWTH, Computer Science 3, many further application-oriented projects based on graph technology, tools, and architectures are and were carried out, in the area of telecommunication, automotive, business administration, civil engineering. eHome, authoring etc. Prof. Nagl has supervised about 300 Masters' Theses and about 50 Doctoral Dissertations. 16 professors were former Ph.D.s of his group. Academic Positions and Honors Prof. Nagl was speaker of Computer Science at RWTH Aachen University from Oct. 2005 to December 2007, after having had various other positions in the department, faculty, and university. He was vice-speaker of Aachen Informatics Industry Club (REGINA) from March 2000 to December 2007, now being a honorary member, and Member of the Board of Forum Informatik from 1988 to 2007 at RWTH Aachen University. He was Head of “Fakultätentag Informatik” (German Informatics Faculty Deans' Conference) from January 2006 to December 2008, and he is now in the board of that association. He was Head of 4ING, the Association of Faculty Conferences of Mechanical Engineering/ Process Engineering, Electrical Engineering/ Information Technology, Civil Engineering/Geodesy, and Informatics in Germany from January 2007 to December 2008, he is now also in the board of that association and a honorary member. He is a Felllow of the Gesellschaft für Informatik, Germanys Association for Computing. He is Speaker Substitute of the Accreditation Committee of EQANIE, European Quality Assurance Network for Informatics Education. He is in the Steering Committee for Faculty Evauations of Informatics Europe. He got a Honorary Doctor from University of Paderborn in 2010.


2011/09/23 Speaker: SOCS Professors
Affiliation: McGill Univerisity
Area: All Computer Science
Title: SOCS Research Overviews
Abstract: This colloquium aims at giving an overview of the research being done within the various groups at SOCS. Several professors are going to present short 5-10 minutes talks about their exciting research. The attendance to the colloquium is mandatory for new M.Sc. students. All students and profs are invited. Biography of Speaker:


2011/09/30 Speaker: Jochen Lang
Affiliation: University of Ottawa
Area: Graphics / Vision / Robotics
Title: Measurement-based Modelling of Haptic Textures
Abstract: Haptic texturing is a widely used and effective technique to increase the realism of virtual reality applications. The sensation of stroking a physical surface with a tool is caused by the geometrical and physical attributes of the surface; attributes that we need to measure and model to improve virtual reality applications. In this presentation, I discuss an interactive and mobile scanning system for the acquisition and synthesis of haptic textures that consists of a visually tracked hand-held touch probe. We discuss two models that we have estimated from these measurements: one model is based on the measured height-profiles of the surface, while the other directly models the accelerations recorded with the probe. We estimate a spatial distribution of infinite-impulse-response filters that operate in the time domain to record the vibratory accelerations. Our results show that the IIR model is effective in representing varying roughness characteristics of both regular-patterned surfaces and stochastic surfaces. Beside roughness, we have also developed a method to estimate the contact stiffness of an object based solely on the acceleration and forces measured during stroking a surface with the hand-held probe. We were able to show an experimental relationship between the estimated stiffness and the contact stiffness observed during compression. We conclude our presentation with a haptic application utilizing our texture models. Biography of Speaker:

Jochen Lang received the M.Sc. degree in computer science from York University, Canada, and the Ph.D. degree in computer science from the University of British Columbia, in 2001. From 2002 to 2004, he was a Postdoctoral Researcher with the Max-Planck-Institut für Informatik, Saarbrücken, Germany. He is currently an Associate Professor with the School of Electrical Engineering and Computer Science, University of Ottawa, where he is a member of the Distributed and Collaborative Virtual Environments Research Laboratory and the Vision, Imaging, Video and Audio Research Laboratory. His research focuses on measurement-based modeling in the areas of computer graphics and haptics. He is working on image-based models, physical models of 3D deformable objects for tracking, and on modelling and rendering for haptics.


2011/10/07 Speaker: Ivan Stojmenovic
Affiliation: University of Ottawa
Title: Contribution of applied algorithms to applied computing
Abstract: There are many attempts to bring together computer scientists, applied mathematician and engineers to discuss advanced computing for scientific, engineering, and practical problems. This talk is about the role and contribution of applied algorithms within applied computing. It will discuss some specific areas where design and analysis of algorithms is believed to be the key ingredient in solving problems, which are often large and complex and cope with tight timing schedules. The talk is based on recent Handbook of Applied Algorithms (Wiley, March 2008), co-edited by the speaker. The featured application areas for algorithms and discrete mathematics include computational biology, computational chemistry, wireless networks, Internet data streams, computer vision, and emergent systems. Techniques identified as important include graph theory, game theory, data mining, evolutionary, combinatorial and cryptographic, routing and localized algorithms. Biography of Speaker:

Ivan Stojmenovic received his Ph.D. degree in mathematics. He held regular and visiting positions in Serbia, Japan, USA, Canada, France, Mexico, Spain, UK (as Chair in Applied Computing at the University of Birmingham), Hong Kong, Brazil, Taiwan, and China, and is Full Professor at the University of Ottawa, Canada and Adjunct Professor at the University of Novi Sad, Serbia. He published over 250 different papers, and edited seven books on wireless, ad hoc, sensor and actuator networks and applied algorithms with Wiley. He is editor of over dozen journals, editor-in-chief of IEEE Transactions on Parallel and Distributed Systems (from January 2010), and founder and editor-in-chief of three journals (MVLSC, IJPEDS and AHSWN). Stojmenovic is one of about 500 computer science researchers with h-index at least 40 and has >11000 citations. He received three best paper awards and the Fast Breaking Paper for October 2003, by Thomson ISI ESI. He is recipient of the Royal Society Research Merit Award, UK. He is elected to IEEE Fellow status (Communications Society, class 2008), and is IEEE CS Distinguished Visitor 2010-12. He received Excellence in Research Award of the University of Ottawa 2009. Stojmenovic chaired and/or organized >60 workshops and conferences, and served in >200 program committees. He was program co-chair at IEEE PIMRC 2008, IEEE AINA-07, IEEE MASS-04&07, EUC-05&08-10, AdHocNow08, IFIP WSAN08, WONS-05, MSN-05&06, ISPA-05&07, founded workshop series at IEEE MASS, ICDCS, DCOSS, WoWMoM, ACM Mobihoc, IEEE/ACM CPSCom, FCST, MSN, and is/was Workshop Chair at IEEE INFOCOM 2011, IEEE MASS-09, ACM Mobihoc-07&08.


2011/10/14 Speaker: Dale Miller
Affiliation: INRIA
Area: Computational logic
Title: Towards a broad spectrum proof certificate
Abstract: Computational logic systems, such as theorem provers and model checkers, produce evidence of a successful proof in an assortment of(often ad hoc) formats. Unfortunately, the evidence generated by one prover is seldom readable by another prover or even by a future version of itself. As a result, provers seldom trust and use proofs from other provers. This situation is made all the more regrettable given that logic and (formal) proof are certainly candidates for universally accepted standards. I will outline some recent work on designing documents, called proof certificates, that satisfy the following requirements: they must be (i) checkable by simple proof checkers, (ii) flexible enough that existing provers can conveniently produce such certificates from their internal evidence of proof, (iii) directly related to proof formalisms used within the structural proof theory literature, and (iv) permit certificates to elide some proof information with the expectation that a proof checker can reconstruct the missing information using bounded and structured proof search. We consider various consequences of these desiderata, including how they can mix computation and deduction and what they mean for the establishment of marketplaces and libraries of proofs. Biography of Speaker:


Dale Miller received his PhD in Mathematics in 1983 from Carnegie Mellon University. He has been a professor at the University of Pennsylvania and Ecole Polytechnique (France) and a Department Head at the Pennsylvania State University. He is currently Director of Research at INRIA-Saclay where he is the Scientific Leader of the INRIA project Parsifal Miller is the Editor-in-Chief of the ACM Transactions on Computational Logic and has editorial duties with the J. of Automated Reasoning, J. of Applied Logic, J. of Logic and Computation, and Theory and Practice of Logic Programming. He has given numerous invited talks at major conferences (eg, CLMPS 2011, APLAS 2010, IJCAR 2006, CSL 2004) as well as numerous invited tutorials (eg, PSL 2011, ISCL 2011, Wollic 2002). He is the recipient of the 2011 LICS Test-of-Time award for a paper he co-authored in LICS 1991. Miller works on many topics in the general area of computational logic, including automated reasoning, logic programming, proof theory, unification theory, operational semantics, and, most recently, proof certificates. He is probably best know for his work on demonstrating computational uses of linear logic and higher-order logic.


2011/10/28 Speaker: Jörg Liebeherr
Affiliation: University of Toronto
Area: Computer Networks
Title: Getting a Grip on Delays in Packet Networks
Abstract: A delay analysis for packet network is notoriously hard. Statistical properties of traffic, link scheduling, and subtle correlations between traffic at different nodes increase the difficulty of characterizing the variable portion of delays. This talk discusses recent progress on the end-to-end delay analysis for a traffic flow in a packet network, using a stochastic network calculus analysis approach. We seek answers to the following questions: What is the relative impact of scheduling and statistical multiplexing on determining delays at a packet switch? What are the scaling properties of end-to-end delays as the number of traversed switches is increased? Does the impact of packet scheduling algorithms diminish when network paths grow large? A key finding is that delays of a flow traversing a sequence of nodes and experiencing cross traffic at each node scale faster than linearly in the number of nodes, when traffic dos not satisfy independence assumptions. More precisely, for exponentially bounded packetized traffic, delays are shown to grow with O(H log H), where H is the number of nodes on the network path. This superlinear scaling of delays is qualitatively different from the scaling behavior predicted by a worst-case analysis or by a probabilistic analysis assuming independence of traffic arrivals. Biography of Speaker:

Jörg Liebeherr received the Ph.D. degree in Computer Science from the Georgia Institute of Technology in 1991. He is currently with the Department of Electrical and Computer Engineering of the University of Toronto as the Nortel Chair of Network Architecture and Services. He is a Fellow of the IEEE. He served on the Board of Governors of the IEEE Communications Society in 2003-2005, and as chair of the IEEE Communications Society Technical Committee on Computer Communications in 2004-2005. He was Editor-in-Chief of IEEE Network in 1999-2000, and an Associate Editor of IEEE/ACM Transactions on Networking and several other journals. He received an NSF Career award in 1996, a best paper award (as co-author) at ACM Sigmetrics 2005, and an Outstanding Service award from the IEEE ComSoc Technical Committee on Computer Communications in 2006. He was co-chair of ACM Sigcomm 2011, which took place in Toronto in August.


2011/11/04 Speaker: Eric Bodden
Affiliation: Technische Universität Darmstadt
Area: Aspect-Orientation
Title: Modular Reasoning in Aspect-Oriented Programs through Join Point Interfaces
Abstract: Important security concerns, such as access control and runtime monitoring can be conveniently implemented using aspect-oriented programming languages. While aspect-oriented programming supports the modular definition of such crosscutting concerns, most approaches to aspect-oriented programming fail, however, to improve, or even preserve, modular reasoning. The main problem is that aspects usually carry, through their pointcuts, explicit references to the base code. These dependencies make programs fragile and hard to reason about. In this talk, we discuss how to separate base code and aspects using Join Point Interfaces, which are contracts between aspects and base code. Base code can define pointcuts that expose selected join points through a Join Point Interface. Conversely, an aspect can offer to advise join points that provide a given Join Point Interface. Crucially, however, aspect themselves cannot contain pointcuts, and hence cannot refer to base code elements. In addition, we will discuss Closure Joinpoints, a mechanism that allows base code to explicitly announce events to aspects. We will discuss why many previous attempts to defining such a language construct fail, and how Closure Joinpoints integrate with Join Point Interfaces to yield a language construct with a clear semantics, avoiding unwanted surprises. Biography of Speaker:

Eric Bodden is currently a research group leader at the European Center for Security and Privacy by Design (EC SPRIDE) and a principal investigator at the Center for Advanced Security Research Darmstadt (CASED). Eric's work focuses on how programming tools and languages can be enhanced to allow programmers to more easily design their programs in a secure way. Eric received his PhD from the McGill School of Computer Science in 2009, under the supervision of Prof. Laurie Hendren, for his work on the Clara framework for partially evaluating AspectJ-based runtime monitors ahead of time.


2011/11/25 Speaker: Emery Berger
Affiliation: Dept. of Computer Science, University of Massachusetts, Amherst
Area: Programming Languages
Title: Multithreaded Programming for Mere Mortals
Abstract: The shift from single to multiple core architectures means that programmers will increasingly be forced to write concurrent, multithreaded programs to increase application performance. Unfortunately, it is challenging to write multithreaded programs that are both correct and fast. This talk presents two software-only systems that aim to dramatically simplify both tasks. The key problem with getting multithreaded programs right is non-determinism. Programs with data races behave differently depending on the vagaries of thread scheduling: different runs of the same multithreaded program can unexpectedly produce different results. These "Heisenbugs" greatly complicate debugging, and eliminating them requires extensive testing to account for possible thread interleavings. We attack the problem of non-determinism with Dthreads, an efficient deterministic multithreading system for general-purpose, unmodified C/C++ programs. Dthreads directly replaces the pthreads library and eliminates races by making all executions deterministic. Not only does Dthreads dramatically outperform a state-of-the-art deterministic runtime system, it often matches—and occasionally exceeds—the performance of pthreads. While correctness is important, it is not enough. Multithreaded applications also need to be efficient and scalable. Key to achieving high performance and scalability is reducing contention for shared resources. However, even when sharing has been reduced to a minimum, threads can still suffer from false sharing. Multiple objects that are not logically shared can end up on the same cache line, leading to invalidation traffic. False sharing is insidious: not only can it be disastrous to performance—causing performance to plummet by as much as an order of magnitude—but it also difficult to diagnose and track down. We have developed two systems to attack the problem of false sharing: Sheriff-Detect and Sheriff-Protect. Sheriff-Detect is a false sharing detection tool that is precise (no false positives), runs with low overhead (on average, 20%), and is accurate, pinpointing the exact objects involved in false sharing. When rewriting a program to fix false sharing is infeasible (source code is unavailable, or padding objects would consume too much memory), programmers can instead use Sheriff-Protect. Sheriff-Protect is a runtime system that automatically eliminates most of the performance impact of false sharing. Sheriff-Protect can improve performance by up to 9X without the need for programmer intervention. Biography of Speaker:

Emery Berger is an Associate Professor in the Department of Computer Science at the University of Massachusetts Amherst. He graduated with a Ph.D. in Computer Science from the University of Texas at Austin in 2002. Professor Berger has been a Visiting Scientist at Microsoft Research and at the Universitat Politècnica de Catalunya (UPC) / Barcelona Supercomputing Center (BSC). Professor Berger's research spans programming languages, runtime systems, and operating systems, with a particular focus on systems that transparently improve reliability and performance. He is the creator of various widely-used software systems including Hoard, a fast and scalable memory manager that accelerates multithreaded applications (and on which the Mac OS X memory manager is based), and DieHard, an error-avoiding memory manager that directly influenced the design of the Windows 7 Fault-Tolerant Heap. His honors include a Microsoft Research Fellowship (2001), an NSF CAREER Award (2003), a Lilly Teaching Fellowship (2006), and a Best Paper Award at FAST 2007. Professor Berger served as the General Chair of the Memory Systems Performance and Correctness workshop (MSPC 2008), Program Chair of the 2010 ACM SIGPLAN/SIGOPS International Conference on Virtual Execution Environments (VEE 2010), and is currently Program Chair for the Workshop on Determinism and Correctness in Parallel Programming, and an Associate Editor of the ACM Transactions on Programming Languages and Systems. He has served on numerous program committees. In his spare time, Professor Berger rides his bicycle, travels to foreign countries, converses in a variety of Romance languages, consumes copious amounts of espresso, and continues his work on a cure for the common cold (which he is certain must somehow involve coffee).


2011/12/16 Speaker: Hans Vangheluwe
Affiliation: University of Antwerp
Area: Software Engineering
Title: Promises and Challenges of Model-Driven Engineering
Abstract: The complexity of (software-intensive) systems we build as well as the demands that are put on quality, safety, and maintainability of these systems has grown drastically over the last decades. To tackle this complexity, Model-Driven Engineering (MDE) treats models, in various formalisms, as first-class artifacts. Such models may be obtained by reverse-engineering of existing software artifacts, for the purpose of analysis, optimization, and evolution. Increasingly, however, software is no longer the primary artifact but rather synthesized from more abstract models. In an attempt to minimize "accidental complexity", the most appropriate modelling languages or formalisms are used for each specific (sub-)problem and phase in the development process. Domain-Specific Modelling (DSM) in particular tries to bridge the gap between the problem domain and the technical solution domain. This has led to a proliferation of the number of (software) modelling languages. This talk will introduce MDE concepts and techniques as well as the research challenges these introduce. Biography of Speaker:

Hans Vangheluwe is a professor in the Department of Mathematics and Computer Science at the University of Antwerp, Belgium and an adjunct professor in the School of Computer Science at McGill University in Montreal, Canada. He holds M.Sc. degrees in theoretical physics and in computer science as well as a D.Sc., all from Ghent University, Belgium. He works on the theoretical foundations as well as on techniques such as meta-modelling and model transformation and tools for modelling and simulation-based design of complex software-intensive systems. Current application domains he works on are automotive, smartphones applications, and Statechart-based modelling, analysis and syntesis of browser-based complex user interfaces. With Juan de Lara, he developed AToM3, A Tool for Multi-formalism and Meta-Modelling, which he uses to prototype new MDE solutions.