Cesar Marcondes

Analysis of TCP live experiments on a real GEO satellite testbed

The paper describes a measurement campaign carried out by the University of Bologna (UoB), the National Inter-University Consortium for Telecommunications (CNIT) and the University of California Los Angeles (UCLA). The aim of the experiments was the performance assessment of a wide range of TCP enhancements on network environments that include a real GEO satellite link. To this end, UoB and CNIT integrated their network tools and set up a testbed composed of a cluster of UoB Linux PCs connected to the CNIT GEO Skyplex satellite platform. Tests were conducted considering both end-to-end TCP enhancements and a TCP-splitting Performance Enhancing Proxy (PEP) developed by UoB. The analysis was not limited to isolated satellite links, but embraced more complex heterogeneous networks, where satellite connections have to compete with wired cross traffic for the network resources. The analysis of the large set of experimental data presented in the paper confirms the challenges posed by GEO satellite channels, which, in this case, were worsened also by the presence of Bandwidth on Demand technique, which impacts TCP performance. Among the end-to-end TCP variants, best results are generally shown by TCP Hybla, especially when heterogeneous environments are considered. On the other hand, the splitting PEP solution offers the best performance with respect to all TCP enhancements, due to its ability to isolate satellite channel impairments. The possible drawback of this approach is in that it violates the end-to-end semantics and is incompatible with IPsec protocol implementation.

A Cross-Comparison of Advanced TCP Protocols in High Speed and Satellite Environments

In todays ever-changing network, new challenges arise with the presence of various types of physical links, such as high speed, satellite, and wireless networks. The common solution is to create a new TCP algorithm that is optimized for each challenge encountered. One example of this is TCP Hybla, which was designed to tackle satellite paths. On the other hand, there are other protocols that claim to be more adaptive to different scenarios, like TCP Adaptive Westwood. In this study, we explore the tradeoffs between utilization and coexistence of both these protocols in two networks: a high-speed and a satellite with identical (and large) Bandwidth-Delay Product. Our goal is to make a comparison of the issues solved by both protocols in both of these scenarios.

JoinUs: Management of Mobile Social Networks for Pervasive Collaboration

Billions of users carry mobile phones nowadays. However, few of them use the phones in more interesting ways than just phone call. In this work, we will present a system design and a proof-of-concept implementation that allows pervasive collaboration among cell phone users using a mobile social network. Our idea is to use new resources available at smart phones to exploit principles of spatial locality (geographical area), temporal locality (instant warning) and interest matching (personal interest on the subject) in order to engage a number of users to collaborate on mini-tasks, also called mini-campaigns that build up spontaneous ldquoflash mobs.

TCP Live Experiments on a Real GEO Satellite Testbed

The paper describes a measurement campaign recently carried out by University of Bologna (UoB), National Inter-University Consortium for Telecommunications-(CNIT) and University of California Los Angeles (UCLA), on the CNIT GEO Skyplex platform. The aim of the experiments is the performance assessment of a wide range of TCP enhancements on network environments that include a real GEO satellite link. Measurements were carried out by means of the integrated UoB-CNIT testbed, fully controlled by remote through a Web interface. Analysis of live experiments data confirm the challenges posed by GEO sat channels, and show a negative influence of the Skyplex BoD technique on TCP start-up performance. Among the TCP enhancements compared in the paper, best results are provided by TCP Hybla and TCP Westwood, with an advantage of the former in heterogeneous environments characterized by a high level of RTT unfairness.

PathCrawler: Automatic harvesting web infra-structure

As network topologies have grown in size and complexity, it is becoming a daunting task for network administrators to keep track the capacity dimensioning of newly installed web-servers within a single or multiple providers. In fact, monitor capacity dimensioning is not a trivial activity since network state changes rather frequently, in particular, in academic environments. In this paper, we describe estimation algorithms and the software architecture of an efficient network management suite to automatically mine path capacity and minimum delays from a venture point to a set of observed web servers. The principle of the suite is based on packet dispersion techniques and repetitive non-intrusive measurements. We provide analytical insights, simulation results and some real case studies where we argument about the correctness, accuracy and usefulness of the suite in the context of management and operation of complex IP based networks.

TCP Adaptive Westwood- Combining TCP Westwood and Adaptive Reno: A Safe Congestion Control Proposal

In this paper, we present the design, implementation and evaluation of a new TCP protocol, TCP-AW (TCP adaptive Westwood). This study was motivated by the intent to address simultaneously several challenging network scenarios, including high bandwidth efficiency in long and fat pipes, RTT fairness, and friendliness to legacy protocol, in a single protocol. TCP-AW leverages the key features of TCP Westwood and TCP adaptive Reno, namely, eligible rate estimation and delay- based adaptive AIMD parameter tuning, respectively. Extensive simulation and measurement results show that TCP AW yields good utilization of available bandwidth and achieves better RTT fairness. As for coexistence with TCP-NewReno, our major safety objective, was that new protocol compares more favorably than other promising protocols such as Hamilton-TCP, CUBIC and Compound-TCP. In contrast to other proposals, TCP-AW has no notions of small and large bandwidth delay product, instead, it scale seamlessly from current Internet paths to faster long distance paths.

TCP in mixed Internet and geo-satellite environments: experiences and results

Experiments in real satellite testbeds are relatively rare in practice. In this paper, we are going to share our experience on using one, and comment on selected sets of results that weve conducted in such mixed Internet and GEO-Satellite environment. The selected material includes: characterization of an induced error-prone satellite channel; results from an optimized non-congestion losses aware TCP (TCP Westwood) and its superior performance compared to TCP NewReno on such error prone scenario. Other results of how implementation details can adversely affect the expected performance, like the congestion window inflation issue and, finally, the state-of-the-art performance enhancement proxies (PEPs) through the satellite. Our main goal is not only to present such novel results but also describe the complex testbed deployment, methodology and the lessons that we learned by performing these experiments

Conservative Slow Start: Controlling Losses in Very High Speed Networks

In this paper, we address the problem of how TCP sessions ramp up their transmission windows in a controlled way. We introduce a conservative slow start scheme that reduces to the regular slow start when session path queues are empty, but slows down the control window increase speed as buffers start to build up. We show via event driven simulation and open source based high speed experimental testbed the effectiveness of our conservative slow start mechanism in reducing packet losses and consequent retransmissions.

Advanced TCP evaluation testbed on mixed wired internet and satellite environments

In this paper, we present our Advanced TCP Evaluation Testbed on Mixed Wired Internet and Satellite Environments.

Shareable Bandwidth Estimation on Satellite Links

Satellite networks have the potentiality to guarantee ubiquitous access to the Internet. Yet, their transmission delay and errors represent still a challenge for researchers and practitioners, especially when considering TCP-based flows. Having efficient means to estimate important channel channel features such as, for instance, the available bandwidth and the total capacity has been proven to be proficient in improving the performance in this context. To this aim we discuss SBE (shareable bandwidth estimator), a solution able to provide a simple and effective estimation of the bottleneck link capacity minus the simultaneously present uniformly distributed traffic. Simulative experiments prove that our scheme is effective since the very beginning of a connection. Furthermore, SBE can be easily implemented as it requires only sender-side modifications and is based on regular TCP functioning.