Cristina Cervelló-Pastor

Prof. Dr. Cristina Cervelló-Pastor

Universitat Politècnica de Catalunya - Barcelona Tech. Department of Network Engineering
orcid

Research Focus & Metrics

Academic Career Length

23Years (2004-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
15 February 2013 in IEEE Transactions on Computers
Verified
Journal Article
Closed access
CPChrysa Papagianni
Chrysa Papagianni
ALAris Leivadeas
Aris Leivadeas
SPSymeon Papavassiliou
Symeon Papavassiliou
VMVassilis Maglaris
Vassilis Maglaris
Cristina Cervelló-PastorCristina Cervelló-Pastor
Cristina Cervelló-Pastor
AMAlvaro Monje
Alvaro Monje
Cloud computing builds upon advances on virtualization and distributed computing to support cost-efficient usage of computing resources, emphasizing on resource scalability and on demand services. Moving away from traditional data-center oriented models, distributed clouds extend over a loosely coupled federated substrate, offering enhanced communication and computational services to target end-users with quality of service (QoS) requirements, as dictated by the future Internet vision. Toward facilitating the efficient realization of such networked computing environments, computing and networking resources need to be jointly treated and optimized. This requires delivery of user-driven sets of virtual resources, dynamically allocated to actual substrate resources within networked clouds, creating the need to revisit resource mapping algorithms and tailor them to a composite virtual resource mapping problem. In this paper, toward providing a unified resource allocation framework for networked clouds, we first formulate the optimal networked cloud mapping problem as a mixed integer programming (MIP) problem, indicating objectives related to cost efficiency of the resource mapping procedure, while abiding by user requests for QoS-aware virtual resources. We subsequently propose a method for the efficient mapping of resource requests onto a shared substrate interconnecting various islands of computing resources, and adopt a heuristic methodology to address the problem. The efficiency of the proposed approach is illustrated in a simulation/emulation environment, that allows for a flexible, structured, and comparative performance evaluation. We conclude by outlining a proof-of-concept realization of our proposed schema, mounted over the European future Internet test-bed FEDERICA, a resource virtualization platform augmented with network and computing facilities.

1 September 2011 in IEEE Communications Surveys & Tutorials
Verified
Journal Article
Closed access
Paulo PereiraPaulo Pereira
Paulo Pereira
ACAugusto Casaca
Augusto Casaca
Joel RodriguesJoel Rodrigues
Joel Rodrigues
Vasco  N. G. J. SoaresVasco N. G. J. Soares
Vasco N. G. J. Soares
JTJoan Triay
Joan Triay
Cristina Cervelló-PastorCristina Cervelló-Pastor
Cristina Cervelló-Pastor
This paper provides an introductory overview of Vehicular Delay-Tolerant Networks. First, an introduction to Delay-Tolerant Networks and Vehicular Delay-Tolerant Networks is given. Delay-Tolerant schemes and protocols can help in situations where network connectivity is sparse or with large variations in density, or even when there is no end-to-end connectivity by providing a communications solution for non real-time applications. Some special issues like routing are addressed in the paper and an introductory description of applications and the most important projects is given. Finally, some research challenges are discussed and conclusions are detailed.

1 June 2014
Verified
Conference Paper
Open access
YJYury Jimenez
Yury Jimenez
Cristina Cervelló-PastorCristina Cervelló-Pastor
Cristina Cervelló-Pastor
AGAurelio J. Garcia
Aurelio J. Garcia
The software-defined network (SDN) advocates a centralized network control, where a controller manages a network from a global view of the network. Large SDN networks may consist of multiple controllers or controller domains that distribute the network management between them, where each controller has a logically centralized but physically distributed vision of the network. In this context, a key challenge faced by providers is to define a scalable control network that exploits the benefits of SDN when used in conjunction with efficient management strategies. Most of the control layer models proposed are not concerned with controller scalability, because they assume that commercial controllers are scalable in terms of capacity (quantity of flows processed per second). However, it has been demonstrated that overloads and long propagation delays among controllers and controllers-switches can lead to a long response time of the controllers, affecting their ability to respond to network events in a very short time and reducing the reliability of communication. In this work we define the principles for designing a scalable control layer for SDN, and show the desired control layer characteristics that optimize the management of the network. We address these principles from the perspective of the controller placement problem. For this purpose we improve and evaluate our previous approach, the algorithm called k-Critical. K-Critical discovers the minimum number of controllers and their location to create a robust control topology that deals robustly with failures and balances the load among the selected controllers. The results demonstrate the effectiveness of our solution by comparing it with other controller placement solutions.

9 July 2020 in Applied Sciences
Verified
Journal Article
Open access
PDF file
MMMiranda McClellan
Miranda McClellan
Cristina Cervelló-PastorCristina Cervelló-Pastor
Cristina Cervelló-Pastor
SSSebastià Sallent-Ribes
Sebastià Sallent-Ribes
Mobile edge computing (MEC) within 5G networks brings the power of cloud computing, storage, and analysis closer to the end user. The increased speeds and reduced delay enable novel applications such as connected vehicles, large-scale IoT, video streaming, and industry robotics. Machine Learning (ML) is leveraged within mobile edge computing to predict changes in demand based on cultural events, natural disasters, or daily commute patterns, and it prepares the network by automatically scaling up network resources as needed. Together, mobile edge computing and ML enable seamless automation of network management to reduce operational costs and enhance user experience. In this paper, we discuss the state of the art for ML within mobile edge computing and the advances needed in automating adaptive resource allocation, mobility modeling, security, and energy efficiency for 5G networks.

1 July 2018 in Journal of Network and Computer Applications
Verified
Journal Article
Open access
ASAlejandro Santoyo-González
Alejandro Santoyo-González
Cristina Cervelló-PastorCristina Cervelló-Pastor
Cristina Cervelló-Pastor

14 September 2019 in Sensors
Verified
Journal Article
Open access
PDF file
ILIrian Leyva-Pupo
Irian Leyva-Pupo
ASAlejandro Santoyo-González
Alejandro Santoyo-González
Cristina Cervelló-PastorCristina Cervelló-Pastor
Cristina Cervelló-Pastor
Achieving less than 1 ms end-to-end communication latency, required for certain 5G services and use cases, is imposing severe technical challenges for the deployment of next-generation networks. To achieve such an ambitious goal, the service infrastructure and User Plane Function (UPF) placement at the network edge, is mandatory. However, this solution implies a substantial increase in deployment and operational costs. To cost-effectively solve this joint placement problem, this paper introduces a framework to jointly address the placement of edge nodes (ENs) and UPFs. Our framework proposal relies on Integer Linear Programming (ILP) and heuristic solutions. The main objective is to determine the ENs and UPFs’ optimal number and locations to minimize overall costs while satisfying the service requirements. To this aim, several parameters and factors are considered, such as capacity, latency, costs and site restrictions. The proposed solutions are evaluated based on different metrics and the obtained results showcase over 20 % cost savings for the service infrastructure deployment. Moreover, the gap between the UPF placement heuristic and the optimal solution is equal to only one UPF in the worst cases, and a computation time reduction of over 35 % is achieved in all the use cases studied.