CNSM 2011 Main Programme
Tuesday, October 25, 2011
08:30AM - 09:00AM Welcome Address
General and TPC Co-chairs
09:00AM - 10:00AM Keynote - Integrating Job Management and Machine Management in Google Data Centers
Joseph Hellerstein, Google Inc, USA
In cloud based services, many jobs run for months. This is a time frame that is on the order of machine lifetimes. As a result, scheduling the administration of machines (e.g., repairs, adding new machines, draining machines) must be integrated with scheduling jobs. At Google, we approach this problem by using an abstraction called resource shapes to describe both machines and jobs. I will discuss the shapes abstraction, and I will also describe some optimization problems that we solve using shapes.
10:00AM - 10:45AM PS1 - Poster Session  
Poster Session Chair: Steven Latre, Ghent University-IBBT, Belgium
A Service Management Architecture Component Model
Jianwen Chen, IBM Australia, Australia.
TARWIS - A Testbed Management Architecture for Wireless Sensor Network Testbeds
Phlipp Hurni, University of Bern, Switzerland
Markus Anwander, University of Berne, Switzerland
Gerald Wagenknecht, Thomas Staub, Torsten Braun, University of Bern, Switzerland.
A Unifying Architecture for Easy Development, Deployment and Management of Voice-Driven Mobile Applications
Jakub Dolezal, Lukas Kencl, Czech Technical University in Prague, Czech Republic.
Collaborative Policy-Based Autonomic Management In a Hierarchical Model
Omid Mola, Michael Bauer, University of Western Ontario, Canada.
An Approach to Peer Selection in Service Overlays
Adriano Fiorese, UDESC, Brazil
Paulo Simoes, Fernando Boavida, University of Coimbra, Portugal.
NEPnet: A Scalable Monitoring System for Anomaly Detection of Network Service
Sujun Cheng, Zhendong Cheng, Zhongzhi Luan, Depei Qian, Beihang University, People Republic of China.
Managing a SaaS Application in the Cloud Using PaaS Policy Sets and a Strategy-Tree
Bradley Simmons, Hamoun Ghanbari, York University, Canada
Marin Litoiu, York Universtity, Canada
Gabriel Iszlai, IBM CAS Toronto, Canada.
A Remote Water Sensing System with Optical Fiber Networks
Yu-Jen Hsu, Chunghwa Telecommunication Laboratories, Taiwan.
10:45AM - 12:15PM ST1- Service Management 
Main Chair: Michele Sibilla, IRIT, France
MODE: Mix Driven On-line Resource Demand Estimation
Amir Kalbasi, University of Calgary, Canada
Diwakar Krishnamurthy, Univ of Calgary, Canada
Jerry Rolia, HP Laboratories, Great Britain
Michael Richter, University of Calgary, Canada.
Performance Management and Quantitative Modeling of IT Service Processes Using Mashup Patterns
Carlos Raniery Paula dos Santos, UFRGS, Brazil
Winnie Cheng, IBM T.J. Watson Research Center, USA
David Loewenstern, IBM TJ Watson Research Center, USA
Larisa Shwartz, IBM T.J. Watson Research, USA
Nikos Anerousis, IBM T.J. Watson Research Center, USA
Lisandro Zambenedetti Granville, UFRGS, Brazil.
AMF Configurations: Checking for Service Protection Using Heuristics
Pejman Salehi, Ferhat Khendek, Concordia University, Canada
Maria Toeroe, Ericsson Inc., Canada
Abdelwahab Hamou-Lhadj, Concordia University, Canada.
2:00PM - 3:30PM      ST2 - Configuration Management
Main Chair: Lisandro Zambenedetti Granville, UFRGS, Brazil
On Synthesizing Distributed Filtering Configuration Considering Risk, Usability and Cost Constraints
Ehab Al-Shaer, University of North Carolina, Charlotte, USA
Bin Zhang, Researcher, USA.
Supporting Vulnerability Awareness in Autonomic Networks and Systems with OVAL
Martin Barrere, LORIA - INRIA Nancy Grand Est, France, France
Remi Badonnel, University of Nancy - LORIA, France
Olivier Festor, INRIA Nancy - Grand Est, France.
Dynamic Service Adaptation for Plug and Play Device Interoperability
Charbel El Kaed, Orangs Labs and University of Grenoble, France
Yves Denneulin, ID/IMAG, France
Francois Gael Ottogalli, Orange Labs, France.
3:30PM - 4:15PM PS2 - Poster Session
Poster Session Chair: Aiko Pras, University of Twente, The Netherlands
Realization of Name Lookup Table in Routers Towards Content-centric Networks
Haesung Hwang, Osaka University, Japan
Shingo Ata, Osaka City University, Japan
Masayuki Murata, Osaka University, Japan.
Control and visualization system for managed self-organization network
Shohei Kamamura, NTT, Japan
Yuki Koizumi, Osaka University, Japan
Takashi Miyamura, NTT, Japan
Shin'ichi Arakawa, Osaka University, Japan
Kohei Shiomoto, NTT, Japan
Masayuki Murata, Osaka University, Japan.
Hotspot Diagnosis on Logical Level
Bo Yang, IBM China Research Lab, PR China
Jeff Kephart, IBM T.J. Watson Research Center, USA
Hendrik Hamann, IBM T.J. Watson Research Center, USA
Stephan Barabasi, IBM, USA.
Introducing Energy-Awareness in Traffic Engineering for Future Networks
George Athanasiou, Kostas Tsagkaris, Panagiotis Vlacheas, Panagiotis Demestichas, University of Piraeus, Greece.
Scalable Root Cause Analysis Assisted by Classified Alarm Information Model Based Algorithm
Masanori Miyazawa, Kosuke Nishimura, KDDI R&D Laboratories Inc., Japan.
Integrated management of network and security devices in IT infrastructures
Bart Vanbrabant, K.U.Leuven, Belgium
Wouter Joosen, KULeuven, Belgium.
Performance Analysis of SNMP in OLSRv2-routed MANETs
Ulrich Herberg, Fujitsu Laboratories of America, France
Robert Cole, US Army CERDEC, USA.
4:15PM - 5:45PM ST3 - Security Management 
Main Chair: Gabi Dreo Rodosek, Universitaet der Bundeswehr Muenchen, Germany
Internet Bad Neighborhods: The Spam Case
Giovane Moura, Ramin Sadre, Aiko Pras, University of Twente, The Netherlands.
Enforcing Security with Behavioral Fingerprinting
Jerome Francois, Radu State, Thomas Engel, University of Luxembourg, Luxemburg
Olivier Festor, INRIA Nancy - Grand Est, France.
Sub-Space Clustering, Inter-Clustering Results Association & Anomaly Correlation for Unsupervised Network Anomaly Detection
Johan Mazel, LAAS-CNRS, France
Pedro Casas, LAAS - Universite de Toulouse, France
Yann Labit, Laboratoire d'Analyse et d'Architecture des Systemes (LAAS-CNRS), France
Philippe Owezarski, LAAS-CNRS, France.
Wednesday, October 26, 2011
09:00AM - 10:00AM Keynote - Services Innovation: Making a difference in the new Information Age
Nikos Anerousis, IBM T.J. Watson Research Center, USA
Organizations that miss a key inflection point or disruption in their industry quickly find themselves at a disadvantage. Drawing from IBM's own history, I show how continuous innovation in products, and more recently services, is a necessity in order to stay competitive in the modern technology landscape. Investing in services research builds strong product ecosystems, relationships with customers, efficient organizations, and raises the quality of life in modern societies. I investigate two  promising areas for research: a) How smart infrastructure and services introduces
limitless possibilities for our neighborhoods and our cities, and, b) how we can redesign our daily work environment for higher efficiency and job satisfaction.
10:00AM - 10:45AM PS3 - Poster Session
Poster Session Chair: Samir Ghamri-Doudane, Alcatel-Lucent Bellabs, France
Universal Engineering Economic Analysis for IT Transformation: Methodology and Tool
Chang-shing Perng, IBM Research, USA
Rong Chang, IBM, USA
Tao Tao, Edward So, IBM Research, USA
Mihwa Choi, IBM Global Services, USA
Hidayatullah Shaikh, IBM, USA.
Autonomic Downlink Inter-Cell Interference Coordination in LTE Self-Organizing Networks
Panagiotis Vlacheas, Evangelos Thomatos, Kostas Tsagkaris, Panagiotis Demestichas, University of Piraeus, Greece.
A Web-based What-If Scenario Analysis Tool for Performance Improvement of IT Support Organizations
Claudio Bartolini, HP Laboratories, USA
Cesare Stefanelli, Davide Targa, Mauro Tortonesi, University of Ferrara, Italy.
Multihop Cellular Network Optimization using genetic algorithms
Velmurugan Ayyadurai, Research Fellow, Great Britain.
Modeling and Simulation of DiffServ Scenarios with the NSDL Framework
Eduardo Marques, Jose J.F. Sousa, University of Madeira, Portugal
Paulo Sampaio, Universidade da Madeira (UMa), Portugal.
A Bi-Criteria Algorithm for Multipoint-to-Multipoint Virtual Connections in Transport Networks
Lucia Martins, University of Coimbra, Portugal
Nuno Ferreira, I.N.E.S.C. Coimbra, Portugal
Jose Craveirinha, Iniversity of Coimbra, Portugal.
Adaptive QoE measurement on videostreaming IP services
Lozano Jose, Telefonica Investigacion y Desarrollo, Spain
Alfonso Castro, Telefonica Research and Development, Spain
Beatriz Fuentes, Telefonica Investigacion y Desarrollo, Spain
Juan Manuel Gonzalez Munoz, Telefonica Global Technology, Spain
Alvaro Rodriguez, Softtelecom S.A., Spain.
10:45AM - 12:15PM ST4 - Self-Managing Networks
Main Chair: Hanan Lutfiyya, University of Western Ontario, Canada
Peer-assisted Network Operator-friendly P2P Traffic Control Technique
HyunYong Lee, NiCT, Japan
Akihiro Nakao, The University of Tokyo, Japan.
Self-adaptive Routing in Multi-hop Sensor Networks
Themistoklis Bourdenas, Imperial College London, Great Britain
David Wood, Petros Zerfos, IBM Research, T.J. Watson, USA
Flavio Bergamaschi, IBM UK Ltd., Great Britain
Morris Sloman, Imperial College London, Great Britain.
Self-Organizing Relays in LTE networks: Queuing analysis and algorithms
Richard Combes, Zwi Altman, Orange Labs, France
Eitan Altman, INRIA, France.
2:00PM - 3:30PM ST5 - Policies 
Main Chair: Michiaki Hayashi, KDDI R&D Laboratories Inc.
Policy-Assisted Planning and Deployment of Virtual Networks
Steven Davy, TSSG, Waterford Institute of Technology, Ireland
Joan Serrat, Universitat Politecnica de Catalunya, Spain
Antonio Astorga, UPC, Barcelona, Spain
Brendan Jennings, TSSG, Waterford Institute of Technology, Ireland
Javier Rubio-Loyola, CINVESTAV Tamaulipas Mexico, Mexico.
Policy Refinement: Decomposition and Operationalization for Dynamic Domains
Robert Craven, Imperial College London, Great Britain
Jorge Lobo, IBM T. J. Watson Research Center, USA
Emil Lupu, Imperial College, Great Britain
Alessandra Russo, Department of Computing, Imperial College London, Great Britain
Morris Sloman, Imperial College London, Great Britain.
Verifying Home Network Bandwidth Sharing Plans
Dimosthenis Pediaditakis, Imperial College of London, Great Britain
Naranker Dulay, Imperial College London, Great Britain.
3:30PM - 4:15PM          PS4 - Poster Session
Poster Session Chair: Yoshiaki Kiriha, NEC, Japan
An autonomic open marketplace for service provisioning and resilience
John Leaney, Haydn Mearns, University of Technology, Sydney, Australia
Artem Parakine, University of Technology Sydney, Australia
John Debenham, University of Technology, Sydney, Australia.
A Trace-Based Service Level Planning Framework for Enterprise Applications Clouds
Anas Youssef, University of Calgary, Canada
Diwakar Krishnamurthy, Univ of Calgary, Canada.
Adaptive Terminal Reporting for Scalable Service Quality Monitoring in Large Networks
Jan Groenendijk, Liam Fallon, Yangcheng Huang, Ericsson, Ireland.
Simulation Based Evaluation of Dynamic Resource Allocation for Adaptive Multimedia Services
Krunoslav Ivesic, Maja Matijasevic, Lea Skorin-Kapov, University of Zagreb, Croatia.
Research on Home NodeB Gateway Load Balancing Mechanism
Xiaojie Wang, LanLan Rui, Yu Peng, Xingyu Chen, Beijing University of Posts and Telecommunications, P.R. China.
QoS based Service Provisioning in NGN/NGS Context
Soumia Kessal, Telecom Paris-Tech, France
Noemie Simoni, Telecom Paris Tech, France.
Unleashing the Power of Policies for Service-Oriented Computing
Carlos Kamienski, Universidade Federal do ABC (UFABC), Brazil
Ramide Dantas, Ernani Azevedo, Cyrus Dias, Universidade Federal de Pernambuco, Brazil
Djamel Sadok, UFPE, Brazil
Borje Ohlman, Ericsson Research, Sweden.
4:15PM - 5:45PM Panel : Bandwidth Shortage and Relevant Solutions: Myths and Facts
Panelists : Prosper Chemouil, Orange Labs, France
Cees de Laat, University of Amsterdam, The Netherlands
George Pavlou, University College London, UK
Aiko Pras, University of Twente, The Netherlands
For a long time Quality of Service (QoS) has been one of the most popular research topics within our network community. The need for QoS is generally motivated by the fact that resources within the Internet are limited and bandwidth therefore scarce. Although such view is shared by many researchers, other researchers claim that optical fiber technologies provide so much capacity, that QoS mechanisms are no longer needed within the core Internet, and over-provisioning will be the way to go.
Discussions between proponents of both approaches or sometimes heated, but not always based in facts. It is the intention of this panel to discuss the facts behind both positions.
Thursday, October 27, 2011
09:00AM - 10:00AM Keynote - Striking the Balance Between Man and Machine in IT Management 
Mark Burgess, CFEngine, Norway
How shall we use technology to manage IT resources? In the past twenty years there have been large shifts in the technologies used to manage IT systems, but very few shifts in the underlying principles. I
believe that we need a better understanding of this collaboration between man and machine to understand management in general. In this keynote, I will use the theme of human knowledge and comprehension to drive a picture of what is needed of IT management in the future.
10:00AM - 10:45AM PS5 - Poster Session 
Poster Session Chair: Filip de Turck, Ghent University-IBBT, Belgium
Automated Service Provisioning and Management for Integrated Femtocell/Wi-Fi and Cellular Networks
Ya-Shian Wang, Chin-Ping Chuang, Kuan-Hsiung Liang, Been-Hwang Liao, Shou-Yi Cheng, Chunghwa Telecom Laboratories, Taiwan.
Towards Green Computing using Diskless High Performance Clusters
Salah Khaled, Khalifa University (KUSTAR), UAE.
Scalable Deterministic End-to-End Probing and Analytical Method for Overlay Network Monitring
Yanjie Ren, Yan Qiao, Xuesong Qiu, Shun-an Wu, Beijing University of Posts and Telecommunications, P.R. China.
A Probe Prediction Approach to Overlay Network Monitoring
Shun-an Wu, Yan Qiao, Xuesong Qiu, Yanjie Ren, Beijing University of Posts and Telecommunications, P.R. China.
Automatic Detection of Sub-optimal Performance in UMTS Networks based on Drive-test Measurements
Oriol Sallent, Jordi Perez-Romero, Universitat Politecnica de Catalunya, Spain
Juan Sanchez-Gonzalez, Universitat Politecnica de Catalunya (UPC), Spain.
Towards Self-Adaptive Management Frameworks: the Case of Aggregated Information Monitoring
Rafik Makhloufi, Universite de Technologie de Troyes, France
Guillaume Doyen, UTT, France
Gregory Bonnet, University of Caen Lower Normandy, France
Dominique Gaiti, Universite de Technologie de Troyes, France.
Towards introspectable, adaptable and extensible autonomic managers.
Yoann Maurel, Laboratoire d'Informatique de Grenoble, France
Philippe Lalanda, Laboratoire Informatique de Grenoble, France
Ada Diaconescu, Telecom ParisTech, France.
10:45AM - 12:15PM ST6 - Change and Workflow Management 
Main Chair: Jose Marcos Nogueira, UFMJ, Brazil
Using Strategy Trees in Change Management in Clouds
Hanan Lutfiyya, University of Western Ontario, Canada
Hiroshi Otsuka, Fujitsu Laboratories Ltd, Japan.
A Performance and Usability Comparison of Automated Planners for IT Change Planning
Sebastian Hagen, Alfons Kemper, Technische Universitaet Muenchen, Germany.
Online Workflow Management and Performance Analysis with STAMPEDE
Dan Gunter, Taghrid Samak, Lawrence Berkeley National Laboratory, USA
Ewa Deelman, ISI/USC, USA
Christopher Brooks, Christopher, USA
Monte Goode, Lawrence Berkeley National Laboratory, USA
Gideon Juve, University of Southern California, USA
Gaurang Mehta, University of Southern California Information Science Institute, Algeria
Priscilla Moraes, University of Delaware, USA.
2:00PM - 3:30PM ST7 - Cloud Management 
Main Chair: Danny Raz, Technion, Israel
CloudOpt: Multi-Goal Optimization of Application Deployments across a Cloud
Jim Zhanwen Li, Carleton University, Canada.
Gossip-based Resource Allocation for Green Computing in Large Clouds
Rerngvit Yanggratoke, Royal Institute of Technology, Sweden
Fetahi Wuhib, KTH Royal Institute of Technology, Sweden
Rolf Stadler, KTH, Sweden.
Service Deactivation Aware Placement and Defragmentation in Enterprise Clouds
Sourav Dutta, IBM Research - India, India
Akshat Verma, IBM India Research Lab, India.
3:30PM - 4:15PM PS6 - Poster Session
Poster Session Chair: Olivier Festor, INRIA, France
Towards a Security Framework for a WS-HumanTask Processor
Sascha Skorupa, Michael Gerhards, Volker Sander, Pascal Pfeiffer, FH Aachen University of Applied Sciences, Germany
Adam Belloum, University of Amsterdam, The Netherlands.
Monitoring network topology dynamism of large-scale traceroute-based measurements
Thomas Bourgeau, Universite Pierre et Marie Curie, France.
Assessing RoQ Attacks on MANETs over Aware and Unaware TPC Techniques
Urlan Barros, UFPR, Brazil
Aldri dos Santos, Michele Nogueira, Universidade Federal do Parana, Brazil.
Functional Components for a Security Manager within Inter-Cloud environments
Michael Kretzschmar, Mario Golling, University of Federal Armed Forces Munich, Germany.
A Unified Approach to Coordinated Energy Management in Data Centers
Rajarshi Das, IBM Research, USA.
Fault Management and Service provisioning process model of next generation access networks
Carmen Mas Machuca, Technische Universitaet Muenchen (TUM), Germany
Sandro Krauss, Deutsche Telekom AG, Germany
Koen Casier, Ghent University - IMEC - IBBT, Belgium.
A Comparison Between Divergence Measures for Network Anomaly Detection
Jean Tajer, University of Paris Descartes, France
Ali Makke, Paris Descartes University, France
Osman Salem, Paris Descartes, France
Ahmed Mehaoua, Paris Descartes University, France.
Reasoning Services for Security and Energy Management in Wireless Sensor Networks
Maissa Mbaye, Universite Gaston Berger, Senegal
Hicham Khalife, LaBRI, Universite de Bordeaux, France
Francine Krief, LaBRI Laboratory, Bordeaux 1 University, France.
4:15PM - 5:45PM ST8 - Service Performance Management
Main Chair: Arthur Andrzejak, University of Heidelberg, Germany
Mitigating the Negative Impact of Preemption On Heterogeneous MapReduce Workloads
Lu Cheng, Qi Zhang, Raouf Boutaba, University of Waterloo, Canada.
Distributed Workload and Response Time Management for Web Applications
Shengzhi Zhang, Penn State University Park, USA
Haishan Wu, Wenjie Wang, IBM Research - China, P.R. China
Bo Yang, IBM China Research Lab, P.R. China
Peng Liu, Penn State University Park, USA.
Staffing Optimization in Complex Service Delivery Systems
Yixin Diao, Aliza Heching, IBM Research, USA.
Friday, October 28, 2011
8:30AM - 9:50AM MC1 - Distributed Management
MiniConference Chair: Hakima Chaouchi, INT-Telecom Sud Paris, France
Outsourced management of home and SOHO desktops
Tiago Cruz, Paulo Simoes, Joao Rodrigues, Edmundo Monteiro, University of Coimbra, Portugal
Fernando Bastos, Alexandre Laranjeira, PT Inovacao, Portugal.
An Ontology-driven Approach to support Wireless Network Monitoring for Home Area Networks
Yuqian Song, John Keeney, Trinity College Dublin, Ireland
Philip Perry, University College Dublin, Ireland
Adriana Hava, UCD, Ireland
Owen Conlan, Trinity College Dublin, Ireland.
Autonomic Management of Workflows on Hybrid Grid-Cloud Infrastructure
Giandomenico Spezzano, CNR-ICAR & University of Calabria, Italy
Giuseppe Papuz, CNR-ICAR, Italy.
Monitoring, Aggregation and Filtering for Efficient Management of Virtual Networks
Stuart Clayman, Richard Clegg, Lefteris Mamatas, George Pavlou, Alex Galis, University College London, Great Britain.
10:20AM - 12:00PM MC2 - Security and Problem Management
MiniConference Chair: Chair: Remi Badonnel, ESIAL - Universit Henri Poincar, Nancy I, France
Witnessing Distributed Denial-of-Service Traffic from an Attacker's Network
Sin-seok Seo, POSTECH, Republic of Korea
Young Won, IIJ, Japan
James Hong, POSTECH, Republic of Korea.
SMURFEN: A Framework of Knowledge Sharing for Collaborative Intrusion Detection
Carol Fung, University of Waterloo, Canada
Quanyan Zhu, University of Illinois at Urbana Champaign, USA
Raouf Boutaba, University of Waterloo, Canada
Tamer Basar, University of Illinios at Urbana-Champaign, USA.
Securing P2P Systems from Sybil Attacks through Adaptive Identity Management
Weverton Luis da Costa Cordeiro, UFRGS, Brazil
Flavio Roberto Santos, Gustavo Mauch, Universidade Federal do Rio Grande do Sul, Brazil
Marinho Barcellos, Luciano Paschoal Gaspary, UFRGS, Brazil.
Semi-Automated Data Center Hotspot Diagnosis
Suzanne McIntosh, Metin Feridun, Axel Tanner, IBM Research, Switzerland
Michael Nidd, IBM Research, Zurich, Switzerland
Jonathan Lenchner, Jeffrey Kephart, IBM, USA
Yang Bo, IBM China Research Laboratory, P.R. China
Stephan Barabasi, IBM, USA.
Cross-Layer Cluster-Based Data Dissemination Failure Detection in MANETS
Hanan Lutfiyya, The University of Western Ontario, Canada
David Kidston, Canada Communications Research Centre,  Canada
Li Li, Canada Communications Research Centre, , Canada
Walee Al Manum, The University of Western Ontario, Canada.
2:00PM - 3:00PM MC3 - Network Embedded Management
MiniConference Chair: Ralf Wolter, CISCO Systems, USA
Increasing Data Center Network Visibility
with Cisco NetFlow-Lite
Luca Deri, ntop, Italy
Ellie Chou, Zach Cherian, Kedar Karmarkar, Cisco, USA
Mike Patterson, Plixer Inc, USA.
Application-defined control of virtual networks over IP-optical networks
Akeo Masuda, Akinori Isogai, Takashi Miyamura, Kohei Shiomoto, Atsushi Hiramatsu, NTT, Japan.
A 100Gig Network Processor platform for Openflow
El Ferkouss Omar, Ilyas Snaiki, UQAM, Canada
Ben Ali Racha, Ericsson, Canada
Yves Lemieux, Ericsson Canada, Canada
Cherkaoui Omar, Professor, Canada.
3:30PM - 5:00PM MC4 - Management Strategies
MiniConference Chair: Stephane Frenot, CITI - INSA de Lyon, France
Criticality Avoidance A new paradigm of congestion control based on the science of phase transition
Siun-Chuon Mau, Akshay Vashist, Alexander Poylisher, Telcordia, USA
Ritu Chadha, Telcordia Technologies, Inc., USA
Chiang Jason, Telcordia, USA.
Towards Decentralized and Adaptive Network Resource Management
Daphne Tuncer, University College of London, Great Britain
Marinos Charalambides, George Pavlou, University College London, Great Britain
Ning Wang, University of Surrey, Great Britain.
Adword auction bidding strategies of budget-limited advertisers on competing search engines
Patrick Maille, Telecom Bretagne, France
Bruno Tuffin, INRIA, France.
Operator Placement with QoS Constraints for Distributed Stream Processing
Yuanqiang Huang, Zhongzhi Luan, Rong He, Depei Qian, Beihang University, People Republic of China.
CNSM'2011 Tutorials
Friday, October 28, 2011
8:30AM - 10:00AM Protocols for Distributed Management
Rolf Stadler, KTH, Sweden
The tutorial covers two fundamental classes of protocols that execute in a distributed management architecture. The first is the class of Echo protocols, which can be used for distributed polling, global state estimation, resource discovery, and distributed configuration. The second class is that of GAP protocols, whose main application is continuous real-time monitoring. All these protocols are based on distributed trees, which are created during execution. Furthermore, they perform in-network aggregation of the results from local operations on network elements. The presentation of the protocols starts with the discussion of their underlying distributed algorithms, which determine key properties and the performance profiles of the protocols.
10:30AM - 12:00AM IT Service Management
Yixin Diao, IBM T. J. Watson Research Center, USA
IT service management includes management processes and technologies that enable service providers to manage the information technology (IT) systems with a focus for delivering business value to customers. The first part of this tutorial overviews the set of best practices used in IT service management processes, which covers the complete lifecycle of IT services including service strategy, service design, service transition, service operation, and continual service improvement. The second part of the tutorial presents several technology considerations from IT Service Providers' perspective on how to manage the service delivery business in order to continually improve service productivity and service quality.
2:00PM - 3:30PM Machine Learning Techniques for Autonomic/Cognitive Networking
Kostas Tsagkaris, Aimilia Bantouna, University of Piraeus, Greece
Cognitive communications rely on past experience and knowledge of the network. This knowledge can be built using user, network and service data combined through machine learning techniques. The latter is split in three major categories with respect to being or not guided by the desired output or rewards when finding the correct answer. These categories are namely: a) the supervised learning techniques, b) the unsupervised learning techniques and c) the reinforcement learning techniques. In particular, the supervised learning techniques are guided by the desired output and thus identify the function that results in that, given the input. Reinforcement learning is respectively guided through the reception of awards or punishments according to the correctness or not of their answer. Finally, unsupervised learning is guided neither by the desired output nor by awards and punishments. On the contrary, they exploit their inputs in such a way so as to identify the pattern that dictates their relations. In this tutorial, an overview of the mathematical foundation of a representative technique from each category is presented and followed by representative applications that are built upon them.
4:00PM - 5:30PM From Supervised to Unsupervised Anomaly Detection in Internet Traffic
Philippe Owezarski, LAAS, France
  The tutorial presents a survey on the state of the art in anomaly detection in Internet traffic and its application to security and traffic management. It covers both the operational aspects of currently running solutions as well as solutions proposed in the literature. As the scope is very wide, a selection of the most significant contributions will be detailed. The tutorial will especially focus on solutions based on machine learning, and especially unsupervised machine learning, and the related solutions able to detect anomalies without any previous knowledge on the traffic. This opens the path to autonomous anomaly detection systems.