• INFRES+LINCS Seminar:
  • When & Where: Seminars are generally (please, check the timeline and location before coming!) held on:
    • Wednesday afternoon at LINCS (14h-15h, Salle de Conseil) and
    • Thursday afternoon at Barrault (14h-15h, Amphi Saphir).
  • Contact us: if you wish to give a talk, please contact

[Next talks] [All talks]

07/01/2015Amal Abdel-Razzak (Orange)Cooperation between LTE and emergent DVB technologies for an efficient delivery of mobile TV
08/01/2015Claudia Malvenuto (Universitą  di Roma La Sapienza)Quasi-symmetric functions, poset partitions and finite topologies
14/01/2015Yanlei Diao (Universityof Massachusetts, Amherst)Platforms and Applications for "Big and Fast" Data Analytics


Date:07/01/2015, 14h
Room:LINCS, Salle de Conseil
Speaker:Amal Abdel-Razzak (Orange)
Talk:Cooperation between LTE and emergent DVB technologies for an efficient delivery of mobile TV
Abstract:The broadcast/cellular cooperation for a common delivery of Mobile TV is at theheart of the emerging mobile broadcast technologies, namely the mobile extensionof the second generation digital video broadcasting for terrestrial reception (DVB-T2Lite) and its follower DVB-Next Generation Handheld (DVB-NGH). These broadcasttechnologies aim to cooperate with the Long Term Evolution (LTE), as the latter isintended to be the bearer of Mobile TV thanks to its enhanced-Multimedia Broadcastand Multicast Service feature (e-MBMS).Even though the 3GPP/DVB cooperation is not a new topic and was investigatedwith the introduction of the previous DVB technology, known as DVB-Handheld(DVB-H), most of the works addressing this issue considered a common service areacovered by both DVB and cellular systems and focused solely on the impact of such cooperation in terms of capacity gains brought by 3GPP and error repair gains broughtby DVB. This strategy was judged to be expensive since a new and very dense DVBnetwork was needed. In order to overcome this problem and decrease as much as possible the need for a new broadcast network, we propose in this thesis a hybrid DVB/LTEnetwork with a coverage extension strategy, where the LTE system, planned for almost a universal coverage, is used to deliver Mobile TV in areas not covered by recentDVB-T2 Lite (or eventually DVB-NGH) network. In this context, we explore twomain issues:1. Mobile TV services have to share LTE resources with other higher priorityservices such as voice traffic. The dynamicity of the latter will impact theQuality of Service (QoS) of Mobile TV. We propose a new QoS-based planning1for the hybrid DVB/LTE so as to guarantee an acceptable watching experiencewithout over-dimensioning the LTE system. We derive using Markov chainanalysis and hitting time theory, several QoS metrics pertaining to mobile TVperformance, such as interruption frequency and duration.2. A new business model which clarifies the relationships between the differentactors of the ecosystem namely DVB and LTE operators as well as the TVchannel providers and constructs the service area from an economic point ofview is needed. In fact, the absence of a clear and viable economic model thatresolves the monetary conflicts between cellular and broadcast operators wasone of the main drawbacks behind the failure of the first attempt of mobile TVdelivery by cooperating UMTS/DVB-H. We develop in this thesis a profit sharing strategy for the cooperative network, using coalition game concept Shapleyvalue and Nash equilibrium for a self-enforcing strategy. We further develop anew framework using real option theory coupled with coalition games for investment decision in mobile TV networks (whether an operator should enterthe mobile TV market and, if yes, when to do so) and show how operators canincorporate the uncertainties related to demand and network operation costs.We propose a bi-level dynamic programming algorithm to solve numerically thedeveloped real option game.
Links:back to the list item, list top or summary top

Date:08/01/2015, 14h30
Room:Salle C48, Barrault
Speaker:Claudia Malvenuto (Universitą  di Roma La Sapienza)
Talk:Quasi-symmetric functions, poset partitions and finite topologies
Abstract:A combinatorial Hopf algebra based on double posets, endowed with abilinear form based on pictures between double posets (in analogy topictures of tableaux as defined by Zelevinski) was introduced in 2011by Malvenuto and Reutenauer.When the second order of a double poset is total, one obtains thenotion of special double poset; it is equivalent to that of labelledposet of Stanley. Its generating function, with respect to Stanley'sclassical definition of P-partitions associated to a special poset Pis quasi-symmetric, and, in fact, it is a homomorphism betweenthe Hopf algebra of double posets and that of quasi-symmetric functions.Generalizing to preorders, we define the notion of T-partitionsassociated to a finite topology T, and deduce a Hopf algebra morphismfrom a new Hopf algebra on topologies to the Hopf algebra of packedwords.
Links:back to the list item, list top or summary top

Date:14/01/2015, 14h
Room:LINCS, Salle du Conseil
Speaker:Yanlei Diao (Universityof Massachusetts, Amherst)
Talk:Platforms and Applications for "Big and Fast" Data Analytics
Abstract:Recently there has been a significant interest in building big data systems that can handle not only "big data" but also "fast data" for analytics. Our work is strongly motivated by recent real-world case studies that point to the need for a general, unified data processing framework to support analytical queries with different latency requirements. Towards this goal, our project is designed to transform the popular MapReduce computation model, originally proposed for batch processing, into distributed (near) real-time processing. In this talk, I start by examining the widely used Hadoop system and presenting a thorough analysis to understand the causes of high latency in Hadoop. I then present a number of necessary architectural changes, as well as new resource configuration and optimization techniques to meet user-specified latency requirements while maximizing throughput. Experiments using typical workloads in click stream analysis and twitter feed analysis show that our techniques reduce the latency from tens or hundreds of seconds in Hadoop to sub-second in our system, with 2x-7x increase in throughput. Our system also outperforms state-of-the-art distributed stream systems, Twitter Storm and Spark Streaming, by a wide margin. Finally, I will show some initial results and challenges of supporting big and fast data analytics in the emerging domain of genomics.
Biography:Yanlei Diao is Associate Professor of Computer Science at the University of Massachusetts Amherst. Her research interests are in information architectures and data management systems, with a focus on big data analytics, data streams, uncertain data management, and RFID and sensor data management. She received her PhD in Computer Science from the University of California, Berkeley in 2005, her M.S. in Computer Science from the Hong Kong University of Science and Technology in 2000, and her B.S. in Computer Science from Fudan University in 1998. Yanlei Diao was a recipient of the 2013 CRA-W Borg Early Career Award (one female computer scientist selected each year), IBM Scalable Innovation Faculty Award, and NSF Career Award, and she was a finalist of the Microsoft Research New Faculty Award. She spoke at the Distinguished Faculty Lecture Series at the University of Texas at Austin. Her PhD dissertation "Query Processing for Large-Scale XML Message Brokering" won the 2006 ACM-SIGMOD Dissertation Award Honorable Mention. She is currently Editor-in-Chief of the ACM SIGMOD Record, Associate Editor of ACM TODS, Area Chair of SIGMOD 2015, and member of the SIGMOD Executive Committee and SIGMOD Software Systems Award Committee. In the past, she has served as Associate Editor of PVLDB, organizing committee member of SIGMOD, CIDR, DMSN, and the New England Database Summit, as well as on the program committees of many international conferences and workshops. Her research has been strongly supported by industry with awards from Google, IBM, Cisco, NEC labs, and the Advanced Cybersecurity Center.
Links:back to the list item, list top or summary top