Network performance | EPFL Graph Search

Network performance refers to measures of service quality of a network as seen by the customer. There are many different ways to measure the performance of a network, as each network is different in nature and design. Performance can also be modeled and simulated instead of measured; one example of this is using state transition diagrams to model queuing performance or to use a Network Simulator. Performance measures The following measures are often considered important:

Bandwidth commonly measured in bits/second is the maximum rate that information can be transferred
Throughput is the actual rate that information is transferred
Latency the delay between the sender and the receiver decoding it, this is mainly a function of the signals travel time, and processing time at any nodes the information traverses
Jitter variation in packet delay at the receiver of the information
Error rate the number of corrupted bits expressed as a percentage or fraction of the total se

Modeling and Measuring Performance of Data Dissemination in Opportunistic Networks

Nikodin Ristanovic

In this thesis we focus on understanding, measuring and describing the performance of Opportunistic Networks (ONs) and their applications. An “opportunistic network” is a term introduced to describe a sparse, wireless, ad hoc network with highly mobile nodes. The opportunistic networking paradigm deviates from the traditional end-to-end connectivity concept: Forwarding is based on intermittent connectivity between mobile nodes (typically, users with wireless devices); complete routes between sources and destinations rarely exist. Due to this unique property of spontaneous link establishment, the challenges that exist in ONs are specific. The unstructured nature of these networks makes it difficult to give any performance guarantees on data dissemination. For this reason, in Part I of this thesis we explore the dynamics that affect the performance of opportunistic networks. We choose a number of meaningful scenarios where our models and algorithms can be validated using large and credible data sets. We show that a drift and jump model that takes a spatial approach succeeds in capturing the impact of infrastructure and mobile-to-mobile exchanges on an opportunistic content update system. We describe the effects of these dynamics by using the age distribution of a dynamic piece of data (i.e., information updates) as the performance measure. The model also succeeds in capturing a strong bias in user mobility and reveals the existence of regions, whose statistics play a critical role in the performance perceived in the network. We exploit these findings to design an application for greedy infrastructure placement, which relies on the model approximation for a large number of nodes. Another great challenge of opportunistic networking lies in the fact that the bandwidth available on wireless links, coupled with ad hoc networking, failed to rival the capacity of backbones and to establish opportunistic networks as an alternative to infrastructure-based networks. For this reason, we never study ONs in an isolated context. Instead, we consider the applications that leverage interconnection between opportunistic networks and legacy networks and we study the benefits this synergy brings to both. Following this approach, we use a large operator-provided data set to show that opportunistic networks (based on Wi-Fi) are capable of offloading a significant amount of traffic from 3G networks. At the same time, the offloading algorithms we propose reduce the amount of energy consumed by mobiles, while requiring Wi-Fi coverage that is several times smaller than in the case of real-time offloading. Again we confirm and reuse the fact that user mobility is biased towards certain regions of the network. In Part II of this thesis, we treat another issue that is essential for the acceptance and evolution of opportunistic networks and their applications. Namely, we address the absence of experimental results that would support the findings of simulation based studies. Although the techniques such as contact-based simulations should intuitively be able to capture the performance of opportunistic applications, this intuition has little evidence in practice. For this reason, we design and deploy an experiment with real users who use an opportunistic Twitter application, in a way that allows them to maintain communication with legacy networks (i.e., cellular networks, the Internet). The experiment gives us a unique insight into certain performance aspects that are typically hidden or misinterpreted when the usual evaluation techniques (such as simulation) are used. We show that, due to the commonly ignored factors (such as the limited transmission bandwidth), contact-based simulations significantly overestimate delivery ratio and obtain delays that are several times lower than those experimentally acquired. In addition to this, our results unanimously show that the common practice of assuming infinite cache sizes in simulation studies, leads to a misinterpretation of the effects of a backbone on an opportunistic network. Such simulations typically overestimate the performance of the opportunistic component, while underestimating the utility of the backbone. Given the discovered deficiencies of the contact-based simulations, we consider an alternative statistical treatment of contact traces that uses the weighted contact graph. We show that this approach offers a better interpretation of the impact of a backbone on an opportunistic network and results in a closer match when it comes to modeling certain aspects of performance (namely, delivery ratio). Finally, the security requirements for the opportunistic applications that involve an interconnection with legacy networks are also highly specific. They cannot be fully addressed by the solutions proposed in the context of autonomous opportunistic (or ad hoc) networks, nor by the security frameworks used for securing the applications with continuous connectivity. Thus, in Part III of this thesis, we put together a security framework that fits the networks and applications that we target (i.e., the opportunistic networks and applications with occasional Internet connectivity). We then focus on the impact of security print on network performance and design a scheme for the protection of optimal relaying capacity in an opportunistic multihop network. We fine-tune the parameters of our scheme by using a game-theoretic approach and we demonstrate the substantial performance gains provided by the scheme.

EPFL2012

Toward Internet Performance Transparency

Georgia Fragkouli

From medical support to education and remote work, our everyday lives increasingly depend on Internet performance. When users experience poor performance, however, the decentralization of the Internet allows limited visibility into which network is responsible. As a result, users are promised Service Level Agreements (SLAs) they cannot verify, regulators make rules they cannot enforce, and networks with competitive performance cannot reliably showcase it to attract new customers. To change this, researchers have proposed transparency protocols, which rely on networks reporting on their own performance. However, these proposals would be hard to adopt because i) they require substantial network resources for extracting and publishing the performance information, or ii) they require cooperative networks that honestly report their performance against their self-interests, or iii) they threaten the anonymizing capability of Tor-like networks by violating their limited visibility assumptions and introducing a new attack vector against them.This dissertation enables network users to estimate the loss and delay of individual networks in an efficient and accurate manner, despite networks generating and controlling the performance data and potentially wanting to exaggerate their performance. It also proposes the first transparency protocol that tries to preserve the capabilities of anonymity networks.The key to efficient and accurate performance monitoring is i) creating incentives for networks to be honest by causing dishonest networks to get into conflict with their neighbors, and ii) combining these incentives with mathematical tools that "tie together" different aspects of network performance.The key to anonymity-preserving monitoring is the insight that users can benefit from transparency even when networks expose coarser-than-per-packet performance information, which at the same time hides sensitive communication patterns and improves anonymity.Our thesis is that efficient and accurate Internet performance transparency is possible and that we can ease the tussle between transparency and user anonymity.

EPFL2022

Energy Savings for Cellular Network with Evaluation of Impact on Data Traffic Performance

Jean-Yves Le Boudec, Byongkwon Moon

We present a concrete methodology for saving energy in future and contemporary cellular networks. It is based on re-arranging the user-cell association so as to allow shutting down under-utilized parts of the network. We consider a hypothetical static case where we have complete knowledge of stationary user locations and thus the results represent an upper bound of potential energy savings. We formulate the problem as a binary integer programming problem, thus it is NP-hard, and we present a heuristic approximation method. We simulate the methodology on an example real cellular network topology with traffic- and user distribution generated according to recently measured patterns. Further, we evaluate the energy savings, using realistic energy profiles, and the impact on the user-perceived network performance, represented by delay and throughput, at various times of day. The general findings conclude that up to 50% energy may be saved in less busy periods, while the performance effects remain limited. We conclude that practical, real-time user-cell re-allocation methodology, taking into account user mobility predictions, may thus be feasible and bring significant energy savings at acceptable performance impact.

2010