GESTIONA:
"Design of Management Strategies for New Network Architectures (Future Internet)" (TEC-34682)

Researchers: Xavi Masip, Eva Marín, Sergi Sánchez,Wilson Ramirez, Anny Martínez, Ghazal Tashakor, Vitor Barbosa, Amir Sinaeepourfard and Shuaib Siddiqui

Duration of the Project: from March 2013 until February 2016

The evolution of the Internet has been remarkable in the last decade, and no limits are foreseen toward what Europe has defined as “Horizon 2020”. Flagship research programmes have been initiated in Europe, United States, and Asia in the last years addressing the design of the network of the future, so called “Future Internet”. This worldwide trend is rooted on the common understanding that current networks must be transformed to accommodate the expected and the unforeseen demands and requirements. Today, two different lines of work exist to conceptually support this transformation. On the one hand, clean slate approaches (also known as revolutionary) push for a new network design that does not necessarily support the current Internet model. On the other hand, evolutionary approaches aim at improving critical aspects of the network without the need of introducing disruptive changes in this latter.
GESTIONA is positioned on this second trend, where the main objective is to contribute to the design of the Future Internet without creating a parallel Internet. While this approach increases the chances of developing products eligible for turning into technology transfer to industry, it also requires innovation on operational networks, which are increasingly heterogeneous and complex to control and manage.
Network operators are currently supporting this network evolution through integration, where traditional telephony networks, TV, public and private networks, are all converging on services offered by a network platform supported by IP over optical network infrastructures. Unfortunately, on the urgency on providing such convergence on the data plane, the Internet community has overlooked other aspects that are critical for the development of the Internet. Three of these key unattended aspects define the pillars in which the GESTIONA proposal will focus on.
GESTIONA proposes to design and validate novel solutions for network management and control leveraging on these three pillars, namely: i) new paradigms supporting coordinated management between layers on multilayer/multi-provider environments; ii) management of programmable networks on open environments; and iii) new strategies for managing routing security in multi-domain scenarios.

   
  "Next Steps in PAth Computation Element (PCE) Architectures: From Software-Defined Concepts to Standards, Interoperability and Deployment"

Researchers: Xavi Masip, Vitor Barbosa, Eva Marín and Wilson Ramirez

Duration of the Project: from November 2013 until October 2015

Consortium: Telefónica I+D (Spain), Technische Universitaet Braunschweig (Germany), Orange Labs, Scuola Sant'Anna (Italy), Old Dog Consulting (UK), Nextworks s.r.l. (Italy), Centre Tecnològic de Telecomunicacions de Catalunya (Spain) and Universitat Politecnica de Catalunya (Spain)

 PCE, a prime example of a network virtualisation and software-defined network artefact, was originally envisaged for constraint-based routing in the conventional best-effort Internet. Over the years, it evolved into the ultimate tool for multi-technology, multidomain, multi-layer, and multi-vendor networking, an effort led by engineers and scientists participating in the PACE consortium. PACE will concentrate European research, development, and standardization, through which innovation is fostered in all areas of technology, network services and broader societal needs. The PACE action will help create and sustain not only a best-of-breed think-tank in this field, but also, and more importantly, a community-led platform for innovation in network engineering and science. The PACE project has the following vision and objectives: 1) Concentrate and rapidly push forward PCE-based frameworks for research, development, technology transfer and innovation, as well as standardisation; 2) Create open-source software, documentation and data repository, as well as sustained collaborative action within a concentrated community of industrial leaders, developers, and academics; and 3) Shape, secure and sustain the EU-based competence, leadership and commitment to innovation into and beyond the era of Software- Defined Networking (SDN), Network Functions Virtualization (NFV) and Application-Based Network Operations (ABNO).

"Towards Automated Interactions between the Internet and the Carrier-Grade Management Ecosystems" (ONE)

Researchers: Marcelo Yannuzzi, Xavi Masip, René Serral and Eva Marín

Duration of the Project: from September 2010 until August 2013

Consortium: Telefónica I+D (Spain), ADVA Optical Networking (Germany), Technische Universitaet Braunschweig (Germany), Seoul National University (South Korea), and Universitat Politecnica de Catalunya (Spain)

The segmentation of IP and carrier-grade technologies has not only produced the carrier's organizational segmentation, but also a fragmentation of the technical competence through separate Network Management Systems (NMSs). In the isolated Internet and carrier-grade management ecosystems, even simple operations, such as IP link upgrade, require multiple human-assisted configurations, and are far from automation. As a result, carriers are seeking ways to alleviate the dependency on manual processes that do not only create management expenditures, but also lead to a heavy overprovision of the IP network. In the project ONE, we propose to break down the current isolation between the IP and carrier-grade management ecosystems. As first step towards a commercially-viable autonomic management solution, we plan to design and prototype an ontology-based communication adapter between the two NMS systems, enabling: i) automated provisioning of IP links; ii) policy-based setup/release of resources; and iii) coordinated self-healing. We emphasize that the solution does not aim to integrate the NMSs, but it should enable their communication, and thus effectively exploit a set of common objectives as they evolve in the current and future systems. 
CRAAX is leading the functional design of the ONE adapter (WP3 Leader) including critical aspects such as Ontology Mapper, Management Controller, and Smart Analytics.

"TEstbed for Future Internet Services" (TEFIS)

Researchers: Marcelo Yannuzzi, Xavi Masip, René Serral, Shuaib Siddiqui

Duration of the Project: from June 2010 until December 2012

Consortium: THALES Services SAS (France), Technische Universitaet Braunschweig (Germany), Engineering Ingegneria Informatica S.p.A. (Italy), Institut National de Recherche en Informatique et Automatique (France), IT Innovation (UK), Fundação de Apoio à Universidade de São Paulo (Brasil), THALES Communications (France), ActiveEon (France), Lulea University of Technology – Centre for Distance-spanning Technology (Sweden), Software Quality System S.A (Spain), Fraunhofer Institute FOKUS (Germany), and Universitat Politecnica de Catalunya (Spain)

CRAAX role in TEFIS includes evaluation of OPENER. We propose two complementary experiments: 1) Quagga OpenAPI Scalability, where we will compare approaches with centralized and distributed node management. 2) IP offloading performance assessment in centralized node management. TEFIS will support Future Internet of Services Research by offering a single access point to different testing and experimental facilities for communities of software and business developers to test, experiment, and collaboratively elaborate knowledge. The project will develop an open platform to access heterogeneous and complementary experimental facilities addressing the full development lifecycle of innovative services with the appropriate tools and testing methodologies. Through the TEFIS platform users will be supported throughout the whole experiment lifecycle by access to different testing tools covering most of the software development-cycle activities such as software build and packaging, compliance tests, system integration, SLA dimensioning, large-scale deployment, and user evaluation of run-time services. The platform will provide the necessary services that will allow the management of underlying testbeds resources. In particular, it will handle generic resource management, resource access scheduling, software deployment, matching and identification of resources that can be activated, and measurement services for a variety of testbeds.

European Union Cost Action IC0806, "Intelligent Monitoring, Control and Security of Critical Infrastructure Systems" (IntelliCIS)

Researchers: Xavi Masip, Marcelo Yannuzzi ,René Serral

Duration of the Project: from 2009 until 2012

Everyday life relies heavily on the reliable operation and intelligent management of large-scale critical infrastructures, such as electric power systems, telecommunication networks, and water distribution networks. The design, monitoring, control and security of such systems are becoming increasingly more challenging as their size, complexity and interactions are steadily growing. Moreover, these critical infrastructures are susceptible to natural disasters, frequent failures, as well as malicious attacks. There is an urgent need to develop a common system-theoretic framework for modelling the behaviour of critical infrastructure systems and for designing algorithms for intelligent monitoring, control and security of such systems. This COST Action (IntelliCIS) will contribute to the advancement of knowledge in the areas of computational intelligence and autonomous agents, with specific emphasis on the application of these methodologies in monitoring and controlling large-scale distributed complex systems. This will be achieved by the development of innovative techniques and algorithms for fault tolerant operation of critical infrastructures and their evaluation by theoretical analysis and simulation. The Action will be a catalyst for instigating interdisciplinary interaction and will promote collaboration between industry, academia and research organizations on the subject of security, quality, reliability, and efficiency of critical infrastructure systems.
Spanish Thematic Network "Future Internet: Eficiencia en las Redes de Altas Prestaciones" (FIERRO), Spanish Ministry of Science and Innovation (TEC2010-12250-E)

Researchers: Xavier Masip

Duration of the Project: from May 2011 to May 2012

FIERRO is a national project which gathers 21 Universities, Research Centers, and Spanish Companies that are carrying out research in the challenging area of the Future Internet, with special focus on high performance IP networks. More specifically, the project is investigating the design of high-speed communication networks, covering the multi-layer aspects and the scalability issues derived from the Internet's growth. More information http://girtel.upct.es/projects/fierro/ (Only in Spanish)
Spanish Thematic Network Multi-layer Networks: IP over Transport Networks" (TEC2008-02552-E/TEC)

Researchers: Xavi Masip

Duration of the Project: from January 2009 to December 2009

Consortium: CTTC, TID, RedIris, i2cat, UPC (2 groups), UPNA, UAM, UC3M, UVA, UPCT, UVI, UdG, and UPV (3 groups).

The goal of this project is the collaboration of different research groups in Spain currently working on the topic of multi-layer IP/optical networks. The project aims to promote the interaction among national researchers, with emphasis on the generation of joint research activities and new knowledge. The project gathers most of the national institutions actively working in the subject of multi-layer transport networks. More Information.

TARIFA: The Atomic Redesign of the Internet Future Architecture
 

Researchers: Xavi Masip

Duration of the Project: from February 2010 to June 2011

Consortium:  UPC (6 groups), Universitat Pompeu Fabra (UPF), Universitat Ramon Llull (URL)

Current Internet architecture has remained nearly invariant since its inception in early 70’s. In incorporating all the services and applications evolution made in almost 40 years, the current architecture has become more and more complex and ossified. Applications, services, networks and protocols are evolving very quickly and it has been widely proved that TCP/IP is not able to efficiently cover the emerging demands of applications, mainly due to its rigid layered model.TARIFA proposes a new clean slate network architecture to deploy ubiquitous services avoiding hierarchical layering. TARIFA follows a service-oriented approach for a flow-oriented context-aware network architecture where communications are composed in situ using reusable and indivisible components called Atomic Services and according to the needs and requirements of the requester.

Sponsored Research Agreement (SRA) with Cisco Systems, Inc., USA

Researchers: Xavi Masip and Marcelo Yannuzzi

Duration of the Project: from July 2010 to May 2011

 

To produce a prototype of a Path-State Vector protocol".

This research program will produce a prototype for testing of the Path-State Vector/BGP overlay developed by CRAAX during the Cisco RFP ''Path-State Vectors (PSVs)'' in 2009. The most important results of the research on PSVs are described in the papers ''Graphs on Path Vectors-Part I: Theory,'' and ''Graphs on Path Vectors-Part II: Integration and Applications'' by Marcelo Yannuzzi, Rene Serral-Gracia, Xavi Masip-Bruin, Fred Baker, Russ White, Pere Monclus, and David Ward (these papers are not public yet). The first purpose of this prototype is to determine if the methods and protocol described in those documents can be used to reduce the BGP convergence time and the churn rate of route advertisements in the interdomain routing system. If successful, other applications shall be developed, integrated, and tested.

Gift granted by the Cisco Collaborative Research Initiative (CCRI)
 

Researchers: Fred Baker, Russ White, David Ward Marcelo Yannuzzi and Xavi Masip

Duration of the Project: from January 2009 to December 2009

Research goal: "Path State Vectors"

Although many arguments can be found in the literature stating that it is time to replace BGP, the practical and economical implications associated with its replacement are obstacles hard to overcome. The challenge nowadays is to find ways to improve different aspects of the inter-domain routing system that neither require the imminent replacement of BGP, nor the development of upward-compatible extensions tending to make BGP even more irreplaceable than it is today. In light of this, this research proposal aims at developing the concept of a Path-State Vector (PSV), as a promising and straightforward way to overlay some key functions out from BGP. In the PSV model that we conceive, the distribution of reachability information (i.e., the localizers) as well as the computation of loop-free paths are kept in the scope of BGP (the underlay), whereas the critical issues currently driving the replacement of BGP can be decoupled from the latter, and managed through one or more domain-level overlays. Unlike pure overlay networks, which simply circumvent BGP, the PSV model that we devise should feed from and assist BGP, as well as offer an effective coupling between BGP and the functions overlayed from the latter. To this end, we propose that PSVs build a graph overlay. Our previous research has shown that graphs inspired in link-state protocols cannot offer a unique and consistent view of the forwarding paths of an internetwork under the current export policies between domains. We have recently found how to build a suitable graph for a PSV, and more importantly, this graph does not violate the policy opaqueness required by ISPs, which provides new and promising research opportunities. As a starting point, this initiative proposes to focus on two objectives: (1) the design of a highly scalable PSV protocol; and (2) exploit the graph overlay in a PSV to remove the path exploration phenomenon from BGP, and therefore, drastically reduce the convergence time on the Internet.

"Diseño Multinivel de Nuevas Arquitecturas y Protocolos para Redes Multidominio" (DOMINO) Spanish Ministry of Science and Innovation (TEC2009-07041)

Researchers: Xavi Masip

Duration of the Project: from January 2010 to December 2012

An essential point while designing the new multidomain network paradigm is to appropriately decide where to allocate and implement the new protocols and features, so as to best fit the expected functionality from this new network architecture. Currently, the physical (optical transport) layer is intelligent enough to endow the network with services that were not conceived some years ago, including enhanced control plane functionalities such as the advertisement of end-to-end routes, QoS routing techniques, etc. This fact indicates that the architectural design must explicitly consider the capabilities and functionality present in both network layers (packet and transport), leading to a joint and multilayer design of the network, so as to optimize both the allocation of functions as well as the required interactions among them. This project aims to elaborate on the multilayer design of a new architecture for multi-domain networks. This architecture will be evaluated and rigorously validated. The main objective of this project is to design a functional architecture focusing on: i) the required mechanisms providing vertical coordination between the packet and transport layers, ii) a new routing paradigm for multidomain networks, addressing the well-known issues in the area, analyzing the potential interaction with the current routing protocol (BGP) in the short-term and even its replacement in a long-term scenario, iii) proposals addressing the overload of the IP addressing scheme semantics, in line with the initiatives recently started by the RRG of the IRTF.

Inter-university cooperation and scientific research program PCI-AECI, "Implementation of Traffic Engineering strategies in countries with poor interdomain connectivity based on the PCE." (A/019977/08)

Researchers: Xavi Masip, Marcelo Yannuzzi

Duration of the Project: from January 2009 to December 2009

This project aims to develop solutions based on the convergence of the PCE and LISP architectures. This line of research is supported by the work done in the first year of cooperation, where the benefits of such convergence were identified. The project has two main objectives: i) the design of a convergent architecture with the LISP data plane completely decoupled from the LISP/PCE control plane; and ii) the design of the intelligent route control algorithms required to dynamically optimize the distribution of traffic for domains with poor inter-domain connectivity, which is the typical situation of many domains in Latin America. More information.
Architectures lab for scalability and interdomain traffic engineering on the Internet: LISP and PCE integration (LITE)

 

Researchers: Xavi Masip, Marcelo Yannuzzi, René Serral

Duration of the Project: from September 2008 to August 2009

One of the proposals to tackle the problems derived from the current addressing scheme is the separation of the location and identification functions inherent in IP addresses. The initiative that is more advanced in terms of specification is LISP (Locator/ID Separation protocol). In LISP, a mapping system is required between the locators and the identifiers. There are currently some proposals for a mapping system for LISP, but none of the solutions under discussion specifies how to apply constraint-based routing techniques to actively engineer interdomain traffic through the mappings. This is especially important for Latin American networks, since these are reachable through heterogenous and typically loaded links, with different characteristics in terms of delays, packet losses, etc. This project aims to extend the implementation of LISP and add constraint-based routing control over the mapping system, with special focus on a convergent PCE/LISP control plane. To this end, the project proposes to start from a running (and tested) implementation of the PCE already developed by the research team. The project will test the extension to LISP using the research infraestructure RedCLARA, as the transport network between two testbeds, one in Spain and the other one in Uruguay.
 Network Equipment for the New Advanced New Advanced Communications Laboratory

Researcher: Xavi Masip

Reference: 20050PEIR0056/15

 This project has funded the laboratory infrastructure initially planned for CRAAX. Specifically, the project financed the purchase of laboratory optical infrastructure (transport layer), including 3 W-onesys ROADMs, 2 CISCO switches (layer 2), and 2 M7 routers (layer 3), and different equipment measurement, hence successfully achieving the initial idea of having a real 3 layers testbed 3 identifying all network layers. No doubt the achievement of this project represented the final impetus for the creation of CRAAX on the Vilanova campus. This allowed the creation of a new stable research group on the Vilanova campus working in the ICT area linked to the Department of Computer Architecture. Researchers currently in CRAAX that participated in the application for this project, want to explicitly mention the support received by UPC Professors, Jordi Domingo and Gabriel Junyent of the departments of Computer Architecture and Signal Theory and Communications respectively . It is also basic to thank the support received by the team Rector, and especially by the then Vice Chancellor for Research, Professor Francesc Xavier Gil, the director of the Vilanova Technology Center, Valenti Guasch and the management team of the Escola Politècnica Superior de Vilanova d'Enginyeria.
 
Colaboration Framework Between The Vilanova I La Geltrú City Council And The Universitat Politecnca De Catalunya Toward R+D+i At Advacned Network Architectures Lab (CRAAX)

Researcher: Xavi Masip

This bilateral agreement between the UPC and the city of Vilanova allowed to obtain the co-financing required for the PEIR project of the Generalitat, and therefore it was essential for the creation of CRAAX. The agreement of 5 anys automatically renewable term and established a clear link between CRAAX and the city of Vilanova and showed the interest and commitment of the city in promoting quality research in the ICT area. It is mandatory to acknowledge the support of the then Mayor Joan Ignasi Helena, and the director of Neapolis, Joan Carles Lluch in achieving the agreement.