Catálogo de publicaciones - libros

This paper presents the idea of a fully decentralized peer-to-peer col-laborative network with a robust, scalable and incentive-compatible system ena-bling storage, retrieval, and manipulation of documents. Modified documents are reviewed in a voting session by other peers. Derived from a scenario, which follows the idea of a distributed encyclopedia, key research questions are concluded.

Xen virtualization is a powerful tool for encapsulating services and providing seamless migration of tasks on hardware failure. This tutorial shows how to set up multiple Xen instances in a network using the MLN management tool.

One of the main challenges in virtual machine administration on a large scale is the specification of complex and repeatable virtualized scenarios. Creating a singe new virtual machine, boot it and install an operating system is straight forward with many of todays tools. But what if you need to deploy 50 identical virtual machines across 25 servers and manage them as an atomic unit? How do you at a later point make consistent design adjustments such as migrating a subset of the virtual machines to a new server or adjusting memory levels? These issues are at the heart of this tutorial.

MLN is a virtual machine management tool supporting both User-Mode Linux and Xen. It has been developed at the Oslo University College in conjunction with its research on system administration and resource management. MLN and its research has previously been presented at the Norwegian Unix User Group (NUUG) and the 20th USENIX Large Installation System Administration conference LISA. This tutorial is interesting for all who want to look beyond the typical one-vm-on-my-desktop scenario. Teachers interested in virtual student labs should also attend.

We start with a short introduction to the Xen virtual machine technology and then proceed to MLN and its own configuration language. In the second part of the tutorial, we will talk about installation and configuration of MLN into a virtual infrastructure across several servers.

Pervasive systems require minimal user intervention while constantly adapting their configuration and behaviour. For example, a personal area network of sensors and computational devices monitoring the health of a patient needs to be able to reconfigure itself in response to sensor changes (failures/removals or additions), changes in the activities or context of the patient as well as changes in the health of the patient.

This tutorial will present the Self-Managed Cell architecture developed at Imperial College in collaboration with the University of Glasgow with par-ticular emphasis on the use of policy-based techniques for implementing adaptation and reconfiguration in autonomous pervasive systems. Policies are rules governing choices in the behaviour of systems and are often en-countered as either event-condition-action rules or authorisation rules, al-though other types of policies such as negotiation, filtering and delegation can be defined. Policies can be used to define management and adaptation behaviours within autonomous cells of devices. In addition, they can govern interactions between and federation of the autonomous cells.

During this tutorial we will present aspects of the Ponder2 policy specification and implementation, structuring concepts for interactions between cells and integration of policy driven interpreters event and domain ser-vices. This tutorial will include a hands-on practical session based on the Ponder2 implementation realised at Imperial College. Ponder2 was recently designed as part of the TrustCoM project and has been used in several projects funded by the European Union and the EPSRC. More information about Ponder2 can be found at http://ponder2.net

This tutorial is aimed at those interested policy-based architectures for autonomous pervasive systems and will allow attendees to gain direct experience with the Ponder2 system. This tutorial is based on a tutorial previously given at the UKUBINET Workshop 2006. Attendees must have a laptop with Java JDK 5.0 installed (Windows, MAC or Linux platform).

Over the past years, telecommunications as a research area has grown far beyond pure communications engineering and today covers the entire economic value chain up to the end customer. As a consequence, “Internet Economics” has been established as a new and promising research area of its own, aiming at a fresh perspective on familiar problems. The basic idea of this interdisciplinary approach is to understand communication networks as economical rather than technical systems, and thus to describe and solve networking issues through the use of economic concepts and techniques. This tutorial gives an introduction into basic notions, concepts and results of this highly interdisciplinary field. We start with summarizing the fundamental framework, focusing on central concepts from operations research, game theory and micro-economics, and introduce basic concepts like equilibria, efficiency, competition models and fairness. This serves as starting point for a comprehensive review of traditional charging schemes like congestion pricing, smart market and Progressive Second-Price auctions, Paris Metro Pricing, effective bandwidth pricing, Cumulus Pricing and the Contract-Balancing Mechanism. The second part of the tutorial starts with a brief overview on relevant charging protocols and architectures, with a special emphasis on the 3GPP IP Multimedia Subsystem (IMS) as a promising candidate for an All-IP Next Generation Network architecture. Finally, we discuss the imminent paradigm shift from charging for Quality-of-Service (QoS) to charging for Quality-of-User Experience (QoE) and present two recent proposals as important examples of current research trends in this stimulating area.

This tutorial explains and demonstrates how the introduction of Semantic Web Services (SWS) to process aware software systems and to Business Process Management (BPM) in general can eliminate the deficiencies that current BPM technology exhibits. The tutorial starts with a thorough discussion of the underlying concepts, Service Oriented Architectures, ontologies, business process management systems and its relevance for today’s system developers. The tutorial will further present the state of the art of current process management systems and Semantic Web Service frameworks. It will motivate the need for explicit use of Semantics to overcome the current weaknesses in BPM, and present a consolidated technical framework that integrates SWS into BPM technology.

The first session will cover the foundations and theoretical aspects, while the second session will be dedicated to a software demonstration and a hands-on session wherein the attendees actively model Business Processes and Semantic Web Services with the respective software tools. Therewith attendees will gain a comprehensive overview of the latest developments in semantically enriched BPM technology, which is one of the central trends in BPM research and development. The tutorial will be held by BPM and SWS experts that actively work on integration of both technologies.

Hard-coding of algorithms with tuneable parameters is not flexible in sensor networks. Also downloading executable files into each sensor node individually might be a problem when single sensor nodes are not permanently reachable or only with high costs. A user should have the possibility for programming a sensor network as a whole in a dynamic way such that the user issues instructions into the sensor network once and code is automatically distributed / executed in the whole sensor network. There are several models for (re)programming wireless sensor networks: the active sensor model based on script interpreters and virtual machines, the mobile agent model, and the database model. The tutorial focuses on the active sensor model and the database model. Reliable transport is essential for transporting management information, configuration data, and code. The tutorial discusses transport protocol design in order to perform error and congestion control.

Over the past years, much effort has been devoted to investigate the interplay of peer-to-peer protocols and the underlying network infrastructure. One surprising result of these studies is the insight that peer-to-peer mechanisms can be pushed down into the network layer. The probably most thoroughly researched embodiment of that idea is the Scalable Source Routing (SSR) protocol. Recently, another similar protocol has been published, the Virtual Ring Routing (VRR) protocol. This tutorial explains both protocols and discusses their potential uses in large self-organizing networks.

SSR and VRR are self-organizing routing protocols which have been inspired by structured peer-to-peer routing protocols such as Chord. Unlike Chord, SSR and VRR are genuine network layer protocols. They are especially suited for networks that do not have a well crafted structure, for example, sensor-actuator networks, ad-hoc networks and mesh networks. In particular, SSR is very memory efficient so that it can provide routing among nodes with very limited resources. This resource efficiency might also prove beneficial for large meshes of multi-homed wireline networks because it greatly reduces the size of the routers’ forwarding information base.

Both, SSR and VRR work on a flat identifier space. As a consequence, they can easily support host mobility without requiring any location directory or other centralized service. Furthermore, SSR and VRR directly provide key based routing in the network layer so that they can serve as efficient basis for fully decentralized applications. Both protocols are based on a virtual ring structure which is used in a Chord-like manner. SSR builds source routes to previously unknown destinations and caches them at so-called intermediate nodes. VRR creates state along the respective paths.

This tutorial gives an in-depth introduction to SSR and VRR. It relates the protocols to previous approaches in ad-hoc networks, mesh networks and sensor networks; and it discusses the various parameters that optimize the protocols in a given scenario. The tutorial concludes with three practical examples from the presenter’s research group. As a result, the participants will be well prepared to apply SSR or VRR for their own purposes.

The IT Infrastructure Library (ITIL) is a today widely-used collection of best practices in IT Service Management that has, of all standardization efforts, gained the biggest popularity. Since it combines the principles of service- and process-orientation in IT Management and is easily accessible, it has become increasingly attractive for IT organizations of almost any size, branch or organizational setup. The scope of ITIL is not limited to technical issues, but also covers the human and economic dimensions (business alignment) of IT Service Management.

In this tutorial we give a survey on the ITIL framework structure and its most important concepts and contents, including an outline of five of ITIL’s core reference processes. Furthermore, the tutorial discusses some important research topics related to ITIL, in particular Management Information Modelling and Tool Support. The five processes selected for presentation within the tutorial are Incident Management, Problem Management, Change Management, Configuration Management and Service Level Management. Learn how theses processes are designed and how they can be implemented.

These topics include the exemplary consideration of an IT incident being recorded, classified and investigated, triggering Problem Management and passing Problem and Error Control before creating a Request for Change for the resolution of the incident and its underlying root cause. Learn how ITIL helps the IT organization to deal with unexpected events in a highly dynamic environment on the one hand, and how it supports the continuous improvement and strategic alignment of IT management. We show how the core processes correlate to each other and point out the central role of Configuration Management and the challenge of setting up a Configuration Management Database (CMDB). Find out what makes a CMDB setup so difficult, which requirements a CMDB should fulfill and why the current commercial and scientific efforts address these challenges insufficiently.

Adequate tools are vital for a successful deployment of ITIL. But since ITIL is tool-independent and hardly formalized, sufficient and integrated tool support for ITIL is not available today. In the tutorial, we present a taxonomy for ITIL processes under tool support aspects by assessing each ITIL process as to its recurrence, lead time, organizational complexity, service level impact and structure.

Finally, we show how ITIL emerged to the ISO/IEC 20000 standard and give an overview of the innovations expected for the next official release “ITIL V.3”, scheduled for this year.

Recently, Steve Jobs, in his public “Thoughts on Music”, pointed out the Digital Rights Management (DRM) systems that Apple has been imposed to adopt for protecting its music against piracy. This brings to a paradox: DRM-protected digital music is prevented from being played by devices of different producers. Conversely, DRM-free content, that uses “open” formats (e.g., MP3 for music and MPEG4 for movies), can be downloaded, distributed, copied and played on different devices. This is an implicit disincentive to legally buy copy-protected digital content, because DRM-free files are interoperable: in fact, 97% of the music filling iPods is unprotected and of obscure origins. Jobs’ conclusions are quite astonishing: abolishing DRMs entirely and selling music encoded in open licensable formats. However, there is no obvious reason for believing that piracy would decrease even if the Steve Jobs’ dream for a “DRM-free” world will finally occur. This implies that future legal market models have to consider serious, scalable, efficient, secure and reliable alternatives to DRM-based on-line (centralized) stores. The Peer-to-Peer paradigm provides a quite mature framework for this applicative domain, making digital content sharing applications a valid solution even for small vendors and emerging artists. In fact, small-medium parties of a market place could hardly afford production and maintenance costs that can be very high if distribution is provided by means of a resilient client-server architecture (e.g., iTunes, Yahoo!, Microsoft Media Shop). But, despite to their big potentials, Peer-to-Peer systems have became infamous through the file sharing applications that make easy for the users to access copy-protected files for free; in fact, it is very difficult to trace the peers’ activity, and identification of abuses cannot be fairly performed because of the absence of a central authority. Moreover, a business model is hard to find: it is questionable if other actors than the of an object should be involved in a transaction as a . The p2p distribution framework leads to technical advantages, but its economical benefits are not clear: the of the bought object can become the distributor later on, but why should he/she provide properly the content if the owner wants to be reimbursed? The tutorial will cover other important services that a market place must include: reputation management, implementation of different preliminary transactions (e.g., bartering, bidding an offer, auctioning), and accounting in decentralized domain. Finally, social networking and self organized communities can be exploited in order to enforce epidemic phenomena and word of mouth marketing. No need to be said, the proper interoperability of incentive strategies, reputation and trust management, accounting solutions, and efficient networking (e.g., search and peer’s cooperation) techniques is critical.