We present SWAN, a system that boosts the utilization of inter-datacenter networks by centrally controlling when and how much traffic each service sends and frequently re-configuring the network's data plane to match current traffic demand. But done simplistically, these re-configurations can also cause severe, transient congestion because different switches may apply updates at different times. We develop a novel technique that leverages a small amount of scratch capacity on links to apply updates in a provably congestion-free manner, without making any assumptions about the order and timing of updates at individual switches. Further, to scale to large networks in the face of limited forwarding table capacity, SWAN greedily selects a small set of entries that can best satisfy current demand. It updates this set without disrupting traffic by leveraging a small amount of scratch capacity in forwarding tables. Experiments using a testbed prototype and data-driven simulations of two production networks show that SWAN carries 60% more traffic than the current practice.

/pdf/achieving-high-utilization-with-software-driven-wan-1eg0vtj3eu.pdf

Achieving high utilization with software-driven WAN

Certain data-communication protocols hog the spotlight, but all of them have a lot in common. Computer Networking: A Top-Down Approach Featuring the Internet explains the engineering problems that are inherent in communicating digital information from point to point. The top-down approach mentioned in the subtitle means that the book starts at the top of the protocol stack--at the application layer--and works its way down through the other layers, until it reaches bare wire. The authors, for the most part, shun the well-known seven-layer Open Systems Interconnection (OSI) protocol stack in favor of their own five-layer (application, transport, network, link, and physical) model. It's an effective approach that helps clear away some of the hand waving traditionally associated with the more obtuse layers in the OSI model. The approach is definitely theoretical--don't look here for instructions on configuring Windows 2000 or a Cisco router--but it's relevant to reality, and should help anyone who needs to understand networking as a programmer, system architect, or even administration guru.The treatment of the network layer, at which routing takes place, is typical of the overall style. In discussing routing, authors James Kurose and Keith Ross explain (by way of lots of clear, definition-packed text) what routing protocols need to do: find the best route to a destination. Then they present the mathematics that determine the best path, show some code that implements those algorithms, and illustrate the logic by using excellent conceptual diagrams. Real-life implementations of the algorithms--including Internet Protocol (both IPv4 and IPv6) and several popular IP routing protocols--help you to make the transition from pure theory to networking technologies. --David WallTopics covered: The theory behind data networks, with thorough discussion of the problems that are posed at each level (the application layer gets plenty of attention). For each layer, there's academic coverage of networking problems and solutions, followed by discussion of real technologies. Special sections deal with network security and transmission of digital multimedia.

Computer Networking: A Top-Down Approach

Netflix is the leading provider of on-demand Internet video streaming in the US and Canada, accounting for 29.7% of the peak downstream traffic in US. Understanding the Netflix architecture and its performance can shed light on how to best optimize its design as well as on the design of similar on-demand streaming services. In this paper, we perform a measurement study of Netflix to uncover its architecture and service strategy. We find that Netflix employs a blend of data centers and Content Delivery Networks (CDNs) for content distribution. We also perform active measurements of the three CDNs employed by Netflix to quantify the video delivery bandwidth available to users across the US. Finally, as improvements to Netflix's current CDN assignment strategy, we propose a measurement-based adaptive CDN selection strategy and a multiple-CDN-based video delivery strategy, and demonstrate their potentials in significantly increasing user's average bandwidth.

Unreeling netflix: Understanding and improving multi-CDN movie delivery

Resource management in a cloud environment is a hard problem, due to: the scale of modern data centers; the heterogeneity of resource types and their interdependencies; the variability and unpredictability of the load; as well as the range of objectives of the different actors in a cloud ecosystem. Consequently, both academia and industry began significant research efforts in this area. In this paper, we survey the recent literature, covering 250+ publications, and highlighting key results. We outline a conceptual framework for cloud resource management and use it to structure the state-of-the-art review. Based on our analysis, we identify five challenges for future investigation. These relate to: providing predictable performance for cloud-hosted applications; achieving global manageability for cloud systems; engineering scalable resource management systems; understanding economic behavior and cloud pricing; and developing solutions for the mobile cloud paradigm .

Resource Management in Clouds: Survey and Research Challenges

To ensure undisrupted business, large Internet companies need to closely monitor various KPIs (e.g., Page Views, number of online users, and number of orders) of its Web applications, to accurately detect anomalies and trigger timely troubleshooting/mitigation. However, anomaly detection for these seasonal KPIs with various patterns and data quality has been a great challenge, especially without labels. In this paper, we proposed Donut, an unsupervised anomaly detection algorithm based on VAE. Thanks to a few of our key techniques, Donut greatly outperforms a state-of-arts supervised ensemble approach and a baseline VAE approach, and its best F-scores range from 0.75 to 0.9 for the studied KPIs from a top global Internet company. We come up with a novel KDE interpretation of reconstruction for Donut, making it the first VAE-based anomaly detection algorithm with solid theoretical explanation.

Unsupervised Anomaly Detection via Variational Auto-Encoder for Seasonal KPIs in Web Applications

We deduce key design features behind the YouTube video delivery system by building a distributed active measurement infrastructure, and collecting and analyzing a large volume of video playback logs, DNS mappings and latency data. We find that the design of YouTube video delivery system consists of three major components: a “flat” video id space, multiple DNS namespaces reflecting a multi-layered logical organization of video servers, and a 3-tier physical cache hierarchy. We also uncover that YouTube employs a set of sophisticated mechanisms to handle video delivery dynamics such as cache misses and load sharing among its distributed cache locations and data centers.

/pdf/vivisecting-youtube-an-active-measurement-study-1qipny71r9.pdf

Vivisecting YouTube: An active measurement study

Effectively managing multiple data centers and their traffic dynamics pose many challenges to their operators, as little is known about the characteristics of inter-data center (D2D) traffic. In this paper we present a first study of D2D traffic characteristics using the anonymized NetFlow datasets collected at the border routers of five major Yahoo! data centers. Our contributions are mainly two-fold: i) we develop novel heuristics to infer the Yahoo! IP addresses and localize their locations from the anonymized NetFlow datasets, and ii) we study and analyze both D2D and client traffic characteristics and the correlations between these two types of traffic. Our study reveals that Yahoo! uses a hierarchical way of deploying data centers, with several satellite data centers distributed in other countries and backbone data centers distributed in US locations. For Yahoo! US data centers, we separate the client-triggered D2D traffic and background D2D traffic from the aggregate D2D traffic using port based correlation, and study their respective characteristics. Our findings shed light on the interplay of multiple data centers and their traffic dynamics within a large content provider, and provide insights to data center designers and operators as well as researchers.

/pdf/a-first-look-at-inter-data-center-traffic-characteristics-bk35077zso.pdf

A first look at inter-data center traffic characteristics via Yahoo! datasets

Using a large Web search service as a case study, we highlight the challenges that modern Web services face in understanding and diagnosing the response time experienced by users. We show that search response time (SRT) varies widely over time and also exhibits counter-intuitive behavior. It is actually higher during off-peak hours, when the query load is lower, than during peak hours. To resolve this paradox and explain SRT variations in general, we develop an analysis framework that separates systemic variations due to periodic changes in service usage and anomalous variations due to unanticipated events such as failures and denial-of-service attacks. We find that systemic SRT variations are primarily caused by systemic changes in aggregate network characteristics, nature of user queries, and browser types. For instance, one reason for higher SRTs during off-peak hours is that during those hours a greater fraction of queries come from slower, mainly-residential networks. We also develop a technique that, by factoring out the impact of such variations, robustly detects and diagnoses performance anomalies in SRT. Deployment experience shows that our technique detects three times more true (operator-verified) anomalies than existing techniques.

/pdf/a-provider-side-view-of-web-search-response-time-3aup5dpsp4.pdf

A provider-side view of web search response time

In this paper we propose VIRO — a novel, virtual identifier (Id) routing paradigm for future networks. The objective is three-fold. First, VIRO directly addresses the challenges faced by the traditional layer-2 technologies such as Ethernet, while retaining its simplicity feature. Second, it provides a uniform convergence layer that integrates and unifies routing and forwarding performed by the traditional layer-2 and layer-3, as prescribed by the traditional local-area/wide-area network dichotomy. Third and perhaps more importantly, VIRO decouples routing from addressing, and thus is namespace-independent. The key idea in our design is to introduce a topology-aware, structured virtual id (vid) space onto which both physical identifiers as well as higher layer addresses/names are mapped. VIRO completely eliminates network-wide flooding in both the data and control planes, and thus is highly scalable and robust. Furthermore, VIRO effectively localizes failures, and possesses built-in mechanisms for fast rerouting and load-balancing.

VIRO: A scalable, robust and namespace independent virtual Id routing for future networks

This paper investigates the roles of front-end (proxy) servers in improving user-perceived performance of dynamic content distribution. Using Bing and Google search services as two case studies, we perform extensive network measurement and analysis to understand several key factors that affect the overall user-perceived performance. In particular, we develop a simple model-based inference framework to indirectly measure and quantify the (directly unobservable) "frontend-to-backend fetching time" comprised of the query processing time at back-end data centers and the delivery time between the back-end data centers and front-end servers. We show that this fetching time plays a critical role in the end-to-end performance of dynamic content delivery.

/pdf/characterizing-roles-of-front-end-servers-in-end-to-end-3stkmom7pa.pdf

Yingying Chen

Papers

Vivisecting YouTube: An active measurement study

A first look at inter-data center traffic characteristics via Yahoo! datasets

A provider-side view of web search response time

VIRO: A scalable, robust and namespace independent virtual Id routing for future networks

Characterizing roles of front-end servers in end-to-end performance of dynamic content distribution