Publicaciones

• “A new heavy-tailed discrete distribution for LRD M/G/∞ sample distribution”
  A. Suárez, J. C. López, C. López, M. Fernández, R. Rodríguez, E. Sousa.
  Performance Evaluation, vol. 47, págs. 197–219, feb. 2002.
  
  Several traffic measurement reports have convincingly shown the presence of self-similarity at the packet level in current networks, inducing as a result a revolution in the stochastic modeling of traffic. The essence of this behavior can be adequately captured by several classes of self-similar stochastic processes. But the use of these in performance analysis has opened new problems and research issues, also in simulation studies where the efficient generation of synthetic sample paths with self-similar properties is one of the fundamental concerns. In this paper, we present a flexible and efficient generator of self-similar traces, based on a simple M/G/∞ model which uses a new heavy-tailed discrete distribution.
• “A Batch Means Procedure for Mean Value Estimation of Processes Exhibiting Long Range Dependence”
  A. Suárez, J. C. López, C. López, M. Rodríguez, M. Veiga, E. Sousa.
  Winter Simulation Conference WSC'02, San Diego (CA), dic. 2002.
  
  Mean value estimation of processes exhibiting Long Range Dependence (LRD) requires a different approach than the techniques applied to those exhibiting Short Range Dependence (SRD),except for the independent replication method. We describe a nonoverlapping batch means method able to deal with LRD processes, the LRD Batch Means method. This method exploits the behavior of Asymptotically Second-order Self-Similar processes: their aggregated processes become well approximated by Fractional Gaussian Noise(FGN) processes for large aggregation levels. Once tested positively this similarity, the method produces a correlation-adjusted confidence interval from an empirical approximation of the distribution of the standardized average for the particular case of FGN processes. Afterwards, we measure its performance over both LRD and SRD processes.
• “A highly efficient M/G/∞ model for generating self-similar traces”
  E. Sousa, A. Suárez, C. López, M. Veiga, J. C. López.
  Winter Simulation Conference WSC'02, San Diego (CA), dic. 2002.
  
  Several traffic measurement reports have convincingly shown the presence of self-similarity in modern networks, inducing as a result a revolution in the stochastic modeling of traffic. The use of self-similar processes in performance analysis has opened new problems and research issues in simulation studies, where the efficient generation of synthetic sample paths with self-similar properties is one of the fundamental concerns. In this paper, we present an M/G/∞ generator of self-similar traces, based on a highly efficient simulation model using the decomposition property of Poisson processes.
• “Simulation Study of Aggregate Flow Control to Improve QoS in a Differentiated Services Network”
  S. Herrería, A. Suárez, M. Veiga, R. F. Rodríguez, C. López
  2nd International Workshop on QoS in Multiservice IP Networks (QoS-IP 2003), Milan (Italia), feb. 2003.
  
  Publicado en: Quality of Service in Multiservice IP Networks,
  M. Ajmone Marsan, G. Corazza, M. Listani, A. Roveri (editores),
  págs. 729–741, Springer, serie LNCS 2601 (ISBN: 3-540-00604-4).
  
  The Differentiated Services architecture is a simple and scalable approach to provide Quality of Service in IP Networks. Several studies have shown that the number of microflows in aggregates, the round trip time or the mean packet size are key factors in the throughput of aggregates obtained using this architecture. In this paper, we examine the behaviour of one of the techniques suggested to improve fairness in a Diffserv network: the Aggregate Flow Control mechanism. We also propose two alternatives in the control overlay of this scheme and compare them with the original approach. Simulation results indicate that our proposed modifications improve throughput assurance and fairness requirements.
• “On the use of control variates in the simulation of queues with heavy-tailed service”
  P. Argibay, A. Suárez, C. López, R. F. Rodríguez, J. C. López, D. Teijeiro
  Industrial Simulation Conference (ISC'03), Valencia (Spain), jun. 2003
  
  Heavy-tailed distributions have been used for modeling several variables involved in telecommunication networks —like page size distributions in Web servers or file size distribution in FTP servers— and in economic models. In G/G/1 queue systems there are few published results to compute their performance parameters analytically, so many times the only method to study them is to use computer simulation. Computer simulation of queue systems where the demanded service time distribution is heavy-tailed presents some difficulties. In this paper we try to gain some insight in those difficulties and evaluate a possible method —based upon control variates— to speed up simulations of G/G/1 queue systems. We present its performance in M/G/1 queue systems, with G heavy-tailed
• “On the simulation of queues with Pareto service”
  P. Argibay, A. Suárez, C. López, R. F. Rodríguez, J. C. López, D. Teijeiro
  ESM'2003, Nottingham (UK), jun. 2003
  
  M/G/n queues —with G a heavy-tailed service time distribution— are used to model queue systems where a range of values of the service time, whose probability is very low, have a drastic impact on the overall performance of the system. The Pareto distribution is one of these heavy-tailed distributions and it has been proposed as the page size distribution in Web servers or as the file size distribution in FTP servers. The accurate analytical treatment of M/G/n systems is very difficult and in many cases it cannot be applied. Simulation is a possible method to study them. But simulations with heavy-tailed random variables present some additional difficulties, and care must be taken when extracting conclusions from the results of these simulations. It is necessary to have accurate and efficient simulators. Efficient because we need to generate big quantities of data for our simulation study to be accurate enough. And the accuracy can be checked by means of comparisons with known results from simpler systems with analytical solution. One of these simpler queue systems that can be studied analytically is the M/P/1 queue. M/P/1 systems can be used then as a workbench for more efficient simulation methods, able to deal with the heavy-tail problematic. The slow convergence of the simulations to the steady state may be an important problem of the simulation of M/P/1 queues. In this paper we investigate this problem and propose a method to try to start the simulation near the steady state, comparing it with the traditional start from empty system.
• “Alternativas para el control de congestión en redes multicast”
  M. Rodríguez, M. Veiga, C. López, S. Herrería
  IV Jornadas de Ingeniería Telemática, Gran Canaria, sep. 2003.
  
  The diffusion of multimedia content across the Internet is still a hot topic of active research. The main difficulty settles in having an efficient transmission protocol that preserves the stability of the current Internet at the same time. In this paper we present our work designing and developing two different protocols aimed at resolving these problems: a multiple layer based one, called LDP and a single-rate one, VLMCC. Both protocols are tuned for the transmission of real-time multimedia content to an unlimited number of receivers. Our simulations show that the protocols are fair against competing TCP traffic present in the network, and have a slowly varying throughput making both protocols suitable for multimedia transmission.
• “RRC-RED: Una solución simple al reparto equitavivo de ancho de banda en RED”
  D. Teijeiro, J. C. López, R. F. Rodríguez, M. Veiga, C. López
  IV Jornadas de Ingeniería Telemática, Gran Canaria, sep. 2003.
  
  One weakness of the RED algorithm typical of routers in the current Internet is that it allows unfair bandwidth sharing when a mixture of traffic types shares a link. This unfairness is caused by the fact that, at any given time, RED imposes the same loss probability on all flows, regardless of their bandwidths. In this paper, we propose Random Rate-Control RED (RRC-RED), a modified version of RED. RRC-RED uses per-active-flow accounting to enforce on each flow a loss rate than depends on the flow s own rate. This papers shows than RRC-RED provides better protection than RED and its variants to solve that problems (like FRED, CHOKe or RED-PD), and, moreover, it is easier to implement and lighter in complexity.
• “On Fair Bandwidth Sharing with RED”
  D. Teijeiro, J. C. López, M. Veiga, C. López, A. Suárez, R. F. Rodríguez, P. Argibay
  18th International Symposium on Computer and Information Sciences (ISCIS'03), Antalya (Turquía), nov. 2003.
  
  Publicado en: Computer and Information Science — ISCIS 2003,
  A. Yazici, C. Sender (editores),
  págs. 651–658, Springer, serie LNCS 2869 (ISBN: 3-540-20409-1).
  
  One weakness of the RED algorithm typical of routers in the current Internet is that it allows unfair bandwidth sharing when a mixture of traffic types shares a link. This unfairness is caused by the fact that, at any given time, RED imposes the same loss probability on all flows, regardless of their bandwidths. In this paper, we propose Random Rate-Control RED (RRC-RED), a modified version of RED. RRC-RED uses per-active-flow accounting to enforce on each flow a loss rate than depends on the flow s own rate. This papers shows than RRC-RED provides better protection than RED and its variants to solve that problems (like FRED, CHOKe or RED-PD), and, moreover, it is easier to implement and lighter in complexity.
• “Effect of the Generation of MPEG-frames within a GOP on Queuing Performance”
  J. C. López, M. Veiga, R. F. Rodríguez, C. López, A. Suárez, D. Teijeiro
  18th International Symposium on Computer and Information Sciences (ISCIS'03), Antalya (Turquía), nov. 2003.
  
  Publicado en: Computer and Information Science — ISCIS 2003,
  A. Yazici, C. Sender (editores),
  págs. 651–658, Springer, serie LNCS 2869 (ISBN: 3-540-20409-1).
  
  MPEG video traffic is expected to represent most of the load in the future high-speed networks. Adequate trafic models for MPEG Variable Bit-Rate video are important for network design, performance evaluation, admission control and resource allocation. It is well-known that VBR video exhibits long-range correlations over arbitrary long time-scales, a phenomenon usually referred to as Long-Range Dependence(LRD). Many models for VBR video traffic have been proposed in the literature. However, while the correlation between Groups of Pictures (GOPs) has been widely analyzed in literature, little effort has been devoted up to now to the generation of the different frame-types within a MPEG GOP. Due to the difficulty and complexity involved in this issue, this is often neglected even though it is a fundamental characteristic of MPEG traffic and it might have an important impact on queuing performance. In fact, many works avoid to descend to the frame lavel proposing models for the GOP process. In this paper, we analyze the impact on queueing performance of different solutions proposed in the literature for generating the different types of MPEG frames within a GOP (these solutions have never been compared and analyzed jointly). In this way, we can be in a good position to use the best model for MPEG in terms of simplicity, computational efficiency and queueing performance, depending on the performance metric to study (loss rate, mean delay and jitter).
• “An Open-Loop Multicast Layered Congestion Protocol for Real-Time Multimedia Transmission”
  M. Rodríguez, M. Veiga, C. López, J. C. López, S. Herrería
  Globecom'03, San Francisco (CA), dic. 2003.
  
  Congestion control of multicast real-time flows in the Internet has different requirements and poses further difficulties than common end-to-end unicast congestion control. In this paper we present a novel congestion control protocol, the LDP, aimed at solving such problems. LDP has been designed to be fair against competing TCP traffic and, at the same time, exhibit smoothness in the instantaneous throughput, making it attractive for real-time multimedia transmission. Furthermore, the open-loop approach makes the protocol highly scalable, as the server load is not affected by the number of clients present in the transmission. Our simulations show that the protocol exhibits the expected TCP fairness while still being highly tunable to smooth the instantaneous throughput.
• “Improving Aggregate Flow Control in Differentiated Services Networks”
  S. Herrería, A. Suárez, M. Veiga, R. F. Rodríguez, C. López
  Computer Networks, vol. 44, núm. 4, págs. 499–512, mar. 2004.
  
  The differentiated services (DiffServ) framework provides a simple and scalable methodology for addressing quality of service issues in IP networks. Service differentiation can be achieved through the assured forwarding (AF) per hop behavior packet-handling scheme. One of the more challenging research issues in DiffServ networks is the fair distribution of bandwidth among aggregates sharing the same AF class. Several studies have shown that the number of microflows in aggregates, the round trip time, the mean packet size and TCP/UDP interactions are key factors in the throughput obtained by aggregates using this architecture. In this paper, we examine the behavior of a recent technique suggested to improve fairness in a DiffServ network: the Aggregate Flow Control mechanism. We also propose two alternatives in the control overlay of this scheme and compare them with the original approach. Simulation results indicate that our proposed modifications improve throughput assurance and fairness requirements.
• “A Receiver Based Single-Layer Multicast Congestion Control Protocol for Multimedia Streaming”
  M. Rodríguez, M. Veiga, S. Herrería, A. Suárez, C. López
  Networking 2004, Atenas, mayo. 2004.
  
  Publicado en: Networking 2004, Networking Technologies, Services and Protocols; Performance of Computer and Communications Networks; Mobile and Wireless Communication,
  N. Mitrou, K. Kontovasilis, I. Iliadir, L. Merekos (editores),
  págs. 550–561, Springer, serie LNCS 3042 (ISBN: 3-540-21959-5).
  
  The transmission of multimedia content in a multicast environment is still a topic of active research. New congestion control protocols that can support the needs of real time transmission, while keeping the current Internet stability, are needed. While there already exist some protocols suitable for multimedia transmission in unicast communications, and there are also protocols capable of transmitting data to a large number of receivers simultaneously, our protocol is designed for achieving both goals at the same time. In this paper we present a novel protocol designed for multimedia streaming in multicast networks. Our simulation results show that the protocol is compatible with TCP flows, thus preserving the network stability, and is suitable for real-time multimedia transmissions, as it has low oscillations in its throughput and imposes no additional network delays that could increase the latency.
• “Square of Service as Control Variate for the Mean Waiting Time Estimation of the M/P/1 Queue”
  A. Suárez, P. Argibay, C. López
  CORS/INFORMS Joint Internation Meeting, Banff (Alberta, Canadá), mayo 2004.
  
  The heavy-tailed condition of the Pareto renders inaccurate most confidence interval methods for the mean queue waiting time of the M/P/1 system. We propose the square of the service time as control variate, which together with bootstrap confidence interval computation gives rise to quite accurate confidence intervals.
• “Improving Fairness Requirements for Assured Services in a Differentiated Services Network”
  S. Herrería, M. Veiga, C. López, M. Rodríguez, A. Suárez
  ICC'04, Paris (Francia), jun. 2004.
  
  One of the more challenging research issues in the context of Assured Forwarding (AF) is the fair distribution of bandwidth among aggregates sharing the same AF class. Several studies have shown that the number of microflows in aggregates, the round trip time, the mean packet size and the TCP/UDP interaction are key factors in the throughput obtained by competing aggregates. Dynamic RIO (DRIO) is an interesting RIO technique suggested to improve assurances among heterogeneous flows in AF-based services. In this paper, we propose applying DRIO to aggregated traffic instead of to individual flows. Working at the aggregate level not only makes DRIO more appropriate to be used in AF networks, but also improves scalability substantially. In addition, we also mitigate some unfairness issues found in the original scheme. Several simulation experiments have been conducted to verify that the proposed improvements allow DRIO to fulfill fairness requirements more satisfactorily.
• “A Generic Approach for the Transmission of Real Time Data in Multicast Networks”
  M. Rodríguez, S. Herrería, M. Veiga, A. Suárez, C. López
  ECUMN'2004, Porto (Portugal), oct. 2004.
  
  We present a new single-rate multicast congestion control protocol aimed at the transmission of real time multimedia data. The protocol extends previous work based on the use of a representative host (leader) decentralising its election. We also introduce an open-loop RTT estimation algorithm that avoids the need to send regular feedback to the sender, and that lets receivers decide to claim for leadership only when needed. Our solution is able to compete against TCP traffic in a fair manner so it can be safely deployed in the Internet. Moreover, we have also made it possible to change the rate adjustment algorithm both at the sender and at the receivers, so that it can be elected according to the application needs or network characteristics.
• “Unbiased RIO: An Active Queue Management Scheme for DiffServ Networks”
  S. Herrería, M. Rodríguez, M. Veiga, A. Suárez, C. López
  ECUMN'2004, Oporto (Portugal), oct. 2004.
  
  One of the more challenging research issues in the context of Assured Forwarding (AF) is the fair distribution of bandwidth among aggregates sharing the same AF class. Several studies have shown that the number of microflows in aggregates, the round trip time, the mean packet size and the TCP/UDP interaction are key factors in the throughput obtained by aggregates using this architecture. We propose the Unbiased RIO (URIO) scheme, an enhanced RIO technique that improves fairness requirements among heterogeneous aggregates in AF-based networks. We validate URIO behaviour by comparing its performance against that of RIO and DRIO through simulation.

Última actualización:
© 2002–2004 Departamento de Ingeniería Telemática, Universidad de Vigo.