Tutorial #1 (Tuesday 16:00-17:30)
Next Generation Network: Networking Virtualization
Prof. Omar Cherkaoui, University of Quebec at Montreal, Canada
Abstract
This tutorial provides an introduction to the discipline of
networking virtualization (NV). Previously, Network
Virtualization has consisted in deploying network services
(VLAN, VPN, etc). One of the emerging concepts is the OS level
of virtualization which consists in injecting multiple instances
of routers on the same network infrastructure. Network
Virtualization is one of the main concepts on which the next
generation network will be based. Many large research projects
(GENI, VINI, VIOLIN) are built on network virtualisation.
The participants will be able to differentiate between the
different virtualization techniques (para-virtualization,
OS-virtualization, application virtualization, service
virtualization, etc.), and the different levels of
virtualization. They will be able to understand the importance
of virtualization and associate the type of virtualisation with
the isolation levels. They will also be able to evaluate the
tradeoffs between performance (latency and bandwidth), security
(isolation) and migration strategy.
One of the main research issues is the sharing of firmware
resources between the different slice routers: processing,
memory, I/O and bandwidth sharing, There are many important
management issues such as configuration, performance and
security (isolation) operations in this new virtualized
environment. The configuration process needs to be enhanced to
include the creation of the virtual routers instances and the
allocation of the appropriate resources for each instance. We
will demonstrate the need for a new performance model over
cross- virtualized network. In terms of security, the need for
new secure operations over this virtualized environment will be
explained.
Instructor Bio
Omar Cherkaoui received his M.Sc. (1981) and Ph.D. (1988) from
the University of Montreal (Canada). He is a Professor of
Computer Science at University of Quebec in Montreal
(Canada). He has (co)authored more than 50 peer-reviewed
technical publications, multiple invited, keynote, and tutorial
presentations, technical reports, and two patent disclosures. He
worked during four years as a researcher at Cisco where he
developed configuration and automatic test case generation
solutions for the GSR. He created the research laboratory in
computer networks (Lab Téléinformatique) where he supervised
more than ten projects in the domain of hi-speed network
management, Web services platform and new multimedia software
(UCLP, Validmaker, Articiel, BIOGRID,..) His research interests
include network management (standardization, protocols,
configuration, validation, modelling, testing), optical
networks, etc. Omar is a member of the technical program
committees of a dozen network management conferences.
Traffic Measurement: Methods, Challenges and Experience
Prof. James Won-ki Hong, POSTECH, Korea
Abstract
The high dynamics of the Internet traffic is making it more
difficult to measure, analyze and understand. However, there is
an increased need for having a good understanding of the
Internet traffic for various applications. This tutorial will
provide various methods used in measuring Internet and
enterprise network traffic. An overview of the related research
activities and research challenges in traffic measurement and
analysis will also be provided. I will then present the R&D
experience on traffic measurement and analysis by POSTECH over
the past 10 years.
Instructor Bio
James Won-Ki Hong (jwkhong@postech.ac.kr) is a professor in the
Dept. of Computer Science and Engineering, POSTECH, Pohang,
Korea. He received a Ph.D. degree from the University of
Waterloo, Canada in 1991 and an M.S. degree from the University
of Western Ontario in 1985. His research interests include
network and systems management, distributed computing, and
network monitoring and analysis, and network planning. James has
served as Technical Chair (1998-2000), Vice Chair (2003-2005)
and Chair (2005-present) for IEEE Comsoc Committee on Network
Operations and Management (CNOM). He has also served as Director
of Online Content for the IEEE Comsoc (Jan. 2004-Dec. 2005). He
is a NOMS/IM Steering Committee Member and a Steering Committee
Member of APNOMS. He was technical co-chair of NOMS 2000 and
APNOMS'99. He was Finance Chair for NOMS 2006, IM 2005 and
Finance Chair and Chair of Local Planning Committee for NOMS
2004. He was the General Chair for APNOMS 2006. He is an
editorial advisory board member of JNSM, IJNM, JTM and TNSM. He
is also editor-in-chief of KNOM Review Journal. He is a member
of IEEE, KICS, KNOM, and KISS. James has previously given
tutorials at NOMS, IM and ICC on the topics of "Web-based
Network Management" and "Internet Network Monitoring and
Analysis".
ISO/IEC 20000: Striving for Resilient IT Services
Thomas Schaaf, MNM Team, University of Munich, Germany
Abstract
Today, IT-driven enterprises all around the world are gearing up
for Service Management using diverse process frameworks like the
IT Infrastructure Library (ITIL) - methods of good practice for
service delivery that were developed by the British government
from the early 1990s and have grown into a ubiquitous code of
practice. Since the release of ISO/IEC 20000 in 2006, an
international standard for IT Service Management is
available. And for the first time, well-defined requirements and
recommendations give guidance and advice in the field of
managing of IT services. More and more organizations are
striving for a third party certification as to ISO/IEC 20000 in
order to give evidence on effective and efficient IT operations
to their customers and stakeholders.
As the world of computing shifts into a service-oriented
paradigm, the scope of ISO/IEC 20000 is not limited to technical
issues, but also covers the human and economic dimensions - "IT
Business-Alignment" (BDIM) of IT Service Management. What are
the tools, methods and philosophies behind this standard? In
this tutorial we run through the basics of processes and service
management.
Instructor Bio
Thomas Schaaf holds a diploma degree in Computer Science and
works as a researcher and teaching associate at the
Ludwig-Maximilians-University of Munich. As a member of the
Munich Network Management (MNM) team, headed by
Prof. H.-G. Hegering, his current research is mainly focused on
Service Level Management tool support issues. Between 2000 and
2005, before joining the MNM team, he worked in Hamburg for AOL
Time Warner. There he was responsible for the development of
online community products and involved into the design of Web
2.0 technologies and features. Thomas Schaaf received the
Foundation, Manager's and Expert Certificate in IT Service
Management according to ITIL/ITILv3 from the Examination
Institute for Information Science (EXIN). In addition, he gives
courses and trainings on IT Service Management as an
EXIN-approved trainer for both ITIL and ISO/IEC 20000. His
recent scientific publications include articles on international
conferences such as DSOM, IM/NOMS/BDIM and HPSUA. In 2007, he
presented a tutorial on AIMS 2007.
Promise Theory - A Practical Introduction
Prof. Mark Burgess, University College Oslo, Norway
Abstract
Most models of computer systems describe information structures,
algorithms or trace event chains in a low level manner. Promise
theory, on the other hand, takes a higher level view that
captures relationships between key system components. This
tutorial offers an introduction to the main concepts of promises
and how to use them to model system behaviour as steady state
equilibrium behaviour.
Promises show us how to incorporate causation, resilience and
also economics into service oriented systems. Using a minimum of
results and several examples, we present a method of analysis
revealing surprises about well-known scenarios.
Instructor Bio
Mark Burgess is professor of Network and System Administration
at Oslo University College. He was the first professor with this
title. Mark obtained a PhD in Theoretical Physics in Newcastle,
for which he received the Runcorn Prize. His current research
interests include the behaviour of computers as dynamic systems
and applying ideas from physics to describe computer
behaviour. Mark is the author of the popular configuration
management software package cfengine. He made important
contributions to the theory of the field of automation and
policy based management, including the idea of operator
convergence and promise theory. He is the author of numerous
books and papers on Network and System Administration and has
won several prizes for his work.
High-speed Passive Packet Capture and Filtering
Luca Deri, University of Pisa, Italy
Abstract
The ntop project was started ten years ago with the goal of
developing an open-source application for passive traffic
monitoring. The first released application, named ntop, was
designed for passive LAN traffic analysis and accessible through
a web interface for immediate user access. Over time ntop has
been adapted to WAN traffic monitoring and today it can be
effectively used for passive packet capture and analysis, as
well as collector for popular protocols such as NetFlow/IPFIX
and sFlow.
In 2001, the author realized that libpcap-based applications
(including ntop) were not powerful enough to keep up with 1 Gbit
networks. Instead of using costly high-speed packet capture
cards, the author designed a Linux kernel module named PF_RING
able to significantly enhance the packet capture speed using
commodity network adapters. Contrary to its initial design,
today PF_RING is used not just for high-speed networking but
also used for speeding-up low-end Linux-based routers.
Recently PF_RING has been recently extended with a plugin
architecture, that allows developers to enhance it with
in-kernel modules for efficient packet capture and analysis.
The first developed plugin targeted VoIP traffic, and it allows
to monitor thousand of simultaneous calls using a low-end PC, in
addition to filter packets at layer 7 contrary to pcap that does
not go beyond packet header.
As happened with 1 Gbit networks, the next monitoring challenge
is 10 Gbit traffic monitoring. Recent innovations in the PC
architecture, paved the way towards a new generation of packet
capture drivers that allow multi-core architecture to be
efficiently exploited. The last part of the tutorial explains
what code modifications are necessary to existing applications
in order to move to efficiently move to 10 Gbit traffic
monitoring.
Instructor Bio
Luca Deri is the leader of the ntop project (http://www.ntop.org/) aimed at
developing an open-source monitoring platform for high-speed
traffic analysis. He worked for University College of London and
IBM Research, prior receiving his PhD at the University of Berne
with a thesis about software components for traffic monitoring
applications. Well known in the open-source and Linux community
as well in the industry where he has been appointed in the
technical advisory board of several leading companies, he
currently shares his time between the ntop project and the
University of Pisa where he has been appointed as lecturer at
the CS department. His home page is http://luca.ntop.org/.
Grid Resource Management
Prof. Joern Altmann, International University, Germany
Abstract
Although Grid computing has been a topic of research for many
years, a truly commercial, sustainable Grid does not exist
yet. One of the reason is the kind of Grid resource management
being used. The aim of this tutorial is to address this issue by
analyzing the current state of the art in Grid architectures and
proposing a new architecture, which is based on an
economic-enhanced Grid resource management.
Besides allowing for different kinds of market mechanisms, the
Grid architecture supports additional services, which enable
companies with little IT knowledge to make use of Grid
computing. These services, which are based on the user's
willingness-to-pay for resources, will make the allocation of
Grid resources easier and the use less risky, enabling a
commercial, sustainable Grid economy. The tutorial will also
discuss the development of such a market-based Grid resource
management by analyzing the technical requirements as well as
the interests of resource providers and consumers. This
interdisciplinary approach to Grid resource management is novel
and will help participants to understand the benefit of using
economic principles for allocating resources.
Instructor Bio
Dr. Jörn Altmann is Associate Professor for Techno-Economics at
Seoul National University and Associate Professor at the
International University of Bruchsal, Germany, where he heads
the group of Computer Networks and Distributed Systems. Prior to
this, he taught computer networks at the University of
California at Berkeley, worked as a Senior Scientist at
Hewlett-Packard Labs, and has been a postdoc at EECS and ICSI of
UC Berkeley. During that time he worked on international
research projects about pricing of network services. Dr. Altmann
received his B.Sc. degree, his M.Sc. degree (1993), and his
Ph.D. (1996) from the University of Erlangen-Nürnberg,
Germany. Dr. Altmann's current research centers on the economics
of Internet services and Internet infrastructures, integrating
economic models into distributed systems.
|
![[AIMS2008]](aims2008.jpg)
![[IFIP]](ifip.png)
![[ACM]](acm.gif)
![[LNCS]](lncs.jpg)
![[EMANICS]](emanics_logo.gif){kind=logo}
![[JACOBS]](jacobs.gif)