Ramesh Johari

End-to-end congestion control for the internet: delays and stability

Under the assumption that queueing delays will eventually become small relative to propagation delays, we derive stability results for a fluid flow model of end-to-end Internet congestion control. The theoretical results of the paper are intended to be decentralized and locally implemented: each end system needs knowledge only of its own round-trip delay. Criteria for local stability and rate of convergence are completely characterized for a single resource, single user system. Stability criteria are also described for networks where all users share the same round-trip delay. Numerical experiments investigate extensions to more general networks. Through simulations, we are able to evaluate the relative importance of queueing delays and propagation delays on network stability. Finally, we suggest how these results may be used to design network resources.

Carving research slices out of your production networks with OpenFlow

1. SLICED PROGRAMMABLE NETWORKS OpenFlow [4] has been demonstrated as a way for researchers to run networking experiments in their production network. Last year, we demonstrated how an OpenFlow controller running on NOX [3] could move VMs seamlessly around an OpenFlow network [1]. While OpenFlow has potential [2] to open control of the network, only one researcher can innovate on the network at a time. What is required is a way to divide, or slice, network resources so that researchers and network administrators can use them in parallel. Network slicing implies that actions in one slice do not negatively affect other slices, even if they share the same underlying physical hardware. A common network slicing technique is VLANs. With VLANs, the administrator partitions the network by switch port and all traffic is mapped to a VLAN by input port or explicit tag. This coarse-grained type of network slicing complicates more interesting experiments such as IP mobility or wireless handover. Here, we demonstrate FlowVisor, a special purpose OpenFlow controller that allows multiple researchers to run experiments safely and independently on the same production OpenFlow network. To motivate FlowVisor’s flexibility, we demonstrate four network slices running in parallel: one slice for the production network and three slices running experimental code (Figure 1). Our demonstration runs on real network hardware deployed on our production network at Stanford and a wide-area test-bed with a mix of wired and wireless technologies.

A buffer-based approach to rate adaptation: evidence from a large video streaming service

Existing ABR algorithms face a significant challenge in estimating future capacity: capacity can vary widely over time, a phenomenon commonly observed in commercial services. In this work, we suggest an alternative approach: rather than presuming that capacity estimation is required, it is perhaps better to begin by using only the buffer, and then ask when capacity estimation is needed. We test the viability of this approach through a series of experiments spanning millions of real users in a commercial service. We start with a simple design which directly chooses the video rate based on the current buffer occupancy. Our own investigation reveals that capacity estimation is unnecessary in steady state; however using simple capacity estimation (based on immediate past throughput) is important during the startup phase, when the buffer itself is growing from empty. This approach allows us to reduce the rebuffer rate by 10-20% compared to Netflixs then-default ABR algorithm, while delivering a similar average video rate, and a higher video rate in steady state.

Efficiency of Scalar-Parameterized Mechanisms

We consider the problem of allocating a fixed amount of an infinitely divisible resource among multiple competing, fully rational users. We study the efficiency guarantees that are possible when we restrict to mechanisms that satisfy certain scalability constraints motivated by large-scale communication networks; in particular, we restrict attention to mechanisms where users are restricted to one-dimensional strategy spaces. We first study the efficiency guarantees possible when the mechanism is not allowed to price differentiate. We study the worst-case efficiency loss (ratio of the utility associated with a Nash equilibrium to the maximum possible utility), and show that Kellys proportional allocation mechanism minimizes the efficiency loss when users are price anticipating. We then turn our attention to mechanisms where price differentiation is permitted; using an adaptation of the Vickrey-Clarke-Groves class of mechanisms, we construct a class of mechanisms with one-dimensional strategy spaces where Nash equilibria are fully efficient. These mechanisms are shown to be fully efficient even in general convex environments, under reasonable assumptions. Our results highlight a fundamental insight in mechanism design: when the pricing flexibility available to the mechanism designer is limited, restricting the strategic flexibility of bidders may actually improve the efficiency guarantee.

A contract-based model for directed network formation

Abstract We consider a network game where the nodes of the network wish to form a graph to route traffic between themselves. We present a model where costs are incurred for routing traffic, as well as for a lack of network connectivity. We focus on directed links and the link stability equilibrium concept, and characterize connected link stable equilibria. The structure of connected link stable networks is analyzed for several special cases.

Downton abbey without the hiccups: buffer-based rate adaptation for HTTP video streaming

Recent work has shown how hard it is to pick a video streaming rate. Video service providers use heuristics to estimate the network capacity leading to unnecessary rebuffering events and suboptimal video quality. This paper argues that we should do away with estimating network capacity, and instead directly observe and control the playback buffer. We present a class of rate selection algorithms that allow us to optimize the delivered video quality while provably never unnecessarily rebuffering. Our algorithms work with discrete video rates, video chunking and for both CBR and VBR video codecs.

Investment and Market Structure in Industries with Congestion

We analyze investment incentives and market structure under oligopoly competition in industries with congestion effects. Our results are particularly focused on models inspired by modern technology-based services such as telecommunications and computing services. We consider situations where firms compete by simultaneously choosing prices and investments; increasing investment reduces the congestion disutility experienced by consumers. We define a notion of returns to investment, according to which congestion models inspired by delay exhibit increasing returns, whereas loss models exhibit nonincreasing returns. For a broad range of models with nonincreasing returns to investment, we characterize and establish uniqueness of pure-strategy Nash equilibrium. We also provide conditions for existence of pure-strategy Nash equilibrium. We extend our analysis to a model in which firms must additionally decide whether to enter the industry. Our theoretical results contribute to the basic understanding of competition in service industries and yield insight into business and policy considerations.

A scalable network resource allocation mechanism with bounded efficiency loss

The design of pricing mechanisms for network resource allocation has two important objectives: 1) a simple and scalable end-to-end implementation and 2) efficiency of the resulting equilibria. Both objectives are met by certain recently proposed mechanisms when users are price taking, but not when users can anticipate the effects of their actions on the resulting prices. In this paper, we partially close this gap, by demonstrating an alternative resource allocation mechanism which is scalable and guarantees a fully efficient allocation when users are price taking. In addition, when links have affine marginal cost, this mechanism has efficiency loss bounded by 1/3 when users are price anticipating. These results are derived by studying Cournot games, and in the process we derive the first nontrivial constant factor bounds on efficiency loss in these well-studied economic models.

Peer-assisted content distribution with prices

Peer-assisted content distribution matches user demand for content with available supply at other peers in the network. Inspired by this supply-and-demand interpretation of the nature of content sharing, we employ price theory to study peer-assisted content distribution. The market-clearing prices are those which align supply and demand, and the system is studied through the characterization of price equilibria. We discuss the efficiency and robustness gains of price-based multilateral exchange, and show that simply maintaining a single price per peer (even across multiple files) suffices to achieve these benefits. Our main contribution is a system design---PACE (Price-Assisted Content Exchange)---that effectively and practically realizes multilateral exchange. Its centerpiece is a market-based mechanism for exchanging currency for desired content, with a single, decentralized price per peer. Honest users are completely shielded from any notion of prices, budgeting, allocation, or other market issues, yet strategic or malicious clients cannot unduly damage the systems efficient operation. Our design encourages sharing of desirable content and network-friendly resource utilization. Bilateral barter-based systems such as BitTorrent have been attractive in large part because of their simplicity. Our research takes a significant step in understanding the efficiency and robustness gains possible with multilateral exchange.

Traffic Engineering vs. Content Distribution: A Game Theoretic Perspective

In this paper we explore the interaction between content distribution and traffic engineering. Because a traffic engineer may be unaware of the structure of content distribution systems or overlay networks, this management of the network does not fully anticipate how traffic might change as a result of his actions. Content distribution systems that assign servers at the application level can respond very rapidly to changes in the routing of the network. Consequently, the traffic engineers decisions may almost never be applied to the intended traffic. We use a game-theoretic framework in which infinitesimal users of a network select the source of content, and the traffic engineer decides how the traffic will route through the network. We formulate a game and prove the existence of equilibria. Additionally, we present a setting in which equilibria are socially optimal, essentially unique, and stable. Conditions under which efficiency loss may be bounded are presented, and the results are extended to the cases of general overlay networks and multiple autonomous systems.