Stefano Traverso

Temporal locality in today's content caching: why it matters and how to model it

The dimensioning of caching systems represents a difficult task in the design of infrastructures for content distribution in the current Internet. This paper addresses the problem of defining a realistic arrival process for the content requests generated by users, due its critical importance for both analytical and simulative evaluations of the performance of caching systems. First, with the aid of \youtube traces collected inside operational residential networks, we identify the characteristics of real traffic that need to be considered or can be safely neglected in order to accurately predict the performance of a cache. Second, we propose a new parsimonious traffic model, named the Shot Noise Model (SNM), that enables users to natively capture the dynamics of content popularity, whilst still being sufficiently simple to be employed effectively for both analytical and scalable simulative studies of caching systems. Finally, our results show that the SNM presents a much better solution to account for the temporal locality observed in real traffic compared to existing approaches.

TailGate: handling long-tail content with a little help from friends

Distributing long-tail content is an inherently difficult task due to the low amortization of bandwidth transfer costs as such content has limited number of views. Two recent trends are making this problem harder. First, the increasing popularity of user-generated content (UGC) and online social networks (OSNs) create and reinforce such popularity distributions. Second, the recent trend of geo-replicating content across multiple PoPs spread around the world, done for improving quality of experience (QoE) for users and for redundancy reasons, can lead to unnecessary bandwidth costs. We build TailGate, a system that exploits social relationships, regularities in read access patterns, and time-zone differences to efficiently and selectively distribute long-tail content across PoPs. We evaluate TailGate using large traces from an OSN and show that it can decrease WAN bandwidth costs by as much as 80% as well as reduce latency, improving QoE. We deploy TailGate on PlanetLab and show that even in the case when imprecise social information is available, TailGate can still decrease the latency for accessing long-tail YouTube videos by a factor of 2.

QoE in Pull Based P2P-TV Systems: Overlay Topology Design Tradeoffs

Abstract-This paper presents a systematic performance analysis of pull P2P video streaming systems for live applications, providing guidelines for the design of the overlay topology and the chunk scheduling algorithm. The contribution of the paper is threefold: 1) we propose a realistic simulative model of the system that represents the effects of access bandwidth heterogeneity, latencies, peculiar characteristics of the video, while still guaranteeing good scalability properties; 2) we propose a new latency/bandwidth-aware overlay topology design strategy that improves application layer performance while reducing the underlying transport network stress; 3) we investigate the impact of chunk scheduling algorithms that explicitly exploit properties of encoded video. Results show that our proposal jointly improves the actual Quality of Experience of users and reduces the cost the transport network has to support.

Unravelling the Impact of Temporal and Geographical Locality in Content Caching Systems

To assess the performance of caching systems, the definition of a proper process describing the content requests generated by users is required. Starting from the analysis of traces of YouTube video requests collected inside operational networks, we identify the characteristics of real traffic that need to be represented and those that instead can be safely neglected. Based on our observations, we introduce a simple, parsimonious traffic model, named shot noise model (SNM), that allows us to capture temporal and geographical locality of content popularity. The SNM is sufficiently simple to be effectively employed in both analytical and scalable simulative studies of caching systems. We demonstrate this by analytically characterizing the performance of the LRU caching policy under the SNM, for both a single cache and a network of caches. With respect to the standard independent reference model (IRM), some paradigmatic shifts, concerning the impact of various traffic characteristics on cache performance, clearly emerge from our results.

Exploring the cloud from passive measurements: The Amazon AWS case

This paper presents a characterization of Amazons Web Services (AWS), the most prominent cloud provider that offers computing, storage, and content delivery platforms. Leveraging passive measurements, we explore the EC2, S3 and CloudFront AWS services to unveil their infrastructure, the pervasiveness of content they host, and their traffic allocation policies. Measurements reveal that most of the content residing on EC2 and S3 is served by one Amazon datacenter, located in Virginia, which appears to be the worst performing one for Italian users. This causes traffic to take long and expensive paths in the network. Since no automatic migration and load-balancing policies are offered by AWS among different locations, content is exposed to the risks of outages. The CloudFront CDN, on the contrary, shows much better performance thanks to the effective cache selection policy that serves 98% of the traffic from the nearest available cache. CloudFront exhibits also dynamic load-balancing policies, in contrast to the static allocation of instances on EC2 and S3. Information presented in this paper will be useful for developers aiming at entrusting AWS to deploy their contents, and for researchers willing to improve cloud design.

The Online Tracking Horde: A View from Passive Measurements

During the visit to any website, the average internaut may face scripts that upload personal information to so called online trackers, invisible third party services that collect information about users and profile them. This is no news, and many works in the past tried to measure the extensiveness of this phenomenon. All of them ran active measurement campaigns via crawlers. In this paper, we observe the phenomenon from a passive angle, to naturally factor the diversity of the Internet and of its users. We analyze a large dataset of passively collected traffic summaries to observe how pervasive online tracking is. We see more than 400 tracking services being contacted by unaware users, of which the top 100 are regularly reached by more than 50 % of Internauts, with top three that are practically impossible to escape. Worse, more than 80 % of users gets in touch the first tracker within 1 second after starting navigating. And we see a lot of websites that hosts hundreds of tracking services. Conversely, those popular web extensions that may improve personal protection, e.g., DoNotTrackMe, are actually installed by a handful of users (3.5 %). The resulting picture witnesses how pervasive the phenomenon is, and calls for an increase of the sensibility of people, researchers and regulators toward privacy in the Internet.

Experimental comparison of neighborhood filtering strategies in unstructured P2P-TV systems

P2P-TV systems performance are driven by the overlay topology that peers form. Several proposals have been made in the past to optimize it, yet little experimental studies have corroborated results. The aim of this work is to provide a comprehensive experimental comparison of different strategies for the construction and maintenance of the overlay topology in P2P-TV systems. To this goal, we have implemented different fully-distributed strategies in a P2P-TV application, called PeerStreamer, that we use to run extensive experimental campaigns in a completely controlled set-up which involves thousands of peers, spanning very different networking scenarios. Results show that the topological properties of the overlay have a deep impact on both user quality of experience and network load. Strategies based solely on random peer selection are greatly outperformed by smart, yet simple strategies that can be implemented with negligible overhead. Even with different and complex scenarios, the neighborhood filtering strategy we devised as most performing guarantees to deliver almost all chunks to all peers with a play-out delay as low as only 6s even with system loads close to 1.0. Results are confirmed by running experiments on PlanetLab. PeerStreamer is open-source to make results reproducible and allow further research by the community.

Unveiling Network and Service Performance Degradation in the Wild with mPlane

Unveiling network and service performance issues in complex and highly decentralized systems such as the Internet is a major challenge. Indeed, the Internet is based on decentralization and diversity. However, its distributed nature leads to operational brittleness and difficulty in identifying the root causes of performance degradation. In such a context, network measurements are a fundamental pillar to shed light on and unveil design and implementation defects. To tackle this fragmentation and visibility problem, we recently conceived mPlane, a distributed measurement platform that runs, collects, and analyzes traffic measurements to study the operation and functioning of the Internet. In this article, we show the potentiality of the mPlane approach to unveil network and service degradation issues in live operational networks, involving both fixed-line and cellular networks. In particular, we combine active and passive measurements to troubleshoot problems in end-customer Internet access connections, or to automatically detect and diagnose anomalies in Internet-scale services (e.g., YouTube) that impact a large number of end users.

Social-Aware Replication in Geo-Diverse Online Systems

Distributing long-tail content is a difficult task due to the low amortization of bandwidth transfer costs as such content has limited number of views. Two recent trends are making this problem harder. First, the increasing popularity of user-generated content and online social networks create and reinforce such popularity distributions. Second, the recent trend of geo-replicating content across multiple points of presence spread around the world, done for improving quality of experience (QoE) for users. In this paper, we analyze and explore the tradeoff involving the “freshness” of the information available to the users and WAN bandwidth costs, and we propose ways to reduce the latter through smart update propagation scheduling, by leveraging on the knowledge of the mapping between social relationships and geographic location, the timing regularities and time differences in end user activity. We first assess the potential of our approach by implementing a simple social-aware scheduling algorithm that operates under bandwidth budget constraints and by quantifying its benefits through a trace-driven analysis. We show that it can reduce WAN traffic by up to 55 percent compared to an immediate update of all replicas, with a minimal effect on information freshness and latency. Second, we build TailGate, a practical system that implements our social-aware scheduling approach, which distributes on the fly long-tail content across PoPs at reduced bandwidth costs by flattening the traffic. We evaluate TailGate by using traces from an OSN and show that it can decrease WAN bandwidth costs by as much as 80 percent and improve QoE. We deploy TailGate on PlanetLab and show that even in the case when imprecise social information is available, it can still decrease by a factor of 2 the latency for accessing long-tail YouTube videos.

Gold Mining in a River of Internet Content Traffic

With the advent of Over-The-Top content providers (OTTs), Internet Service Providers (ISPs) saw their portfolio of services shrink to the low margin role of data transporters. In order to counter this effect, some ISPs started to follow big OTTs like Facebook and Google in trying to turn their data into a valuable asset. In this paper, we explore the questions of what meaningful information can be extracted from network data, and what interesting insights it can provide. To this end, we tackle the first challenge of detecting “user-URLs”, i.e., those links that were clicked by users as opposed to those objects automatically downloaded by browsers and applications. We devise algorithms to pinpoint such URLs, and validate them on manually collected ground truth traces. We then apply them on a three-day long traffic trace spanning more than 19,000 residential users that generated around 190 million HTTP transactions. We find that only 1.6% of these observed URLs were actually clicked by users. As a first application for our methods, we answer the question of which platforms participate most in promoting the Internet content. Surprisingly, we find that, despite its notoriety, only 11% of the user URL visits are coming from Google Search.