Given the technical flaws with---and the increasing non-observance of---the TCP-friendliness paradigm, we must rethink how the Internet should manage bandwidth allocation. We explore this question from first principles, but remain within the constraints of the Internet's current architecture and commercial arrangements. We propose a new framework, Recursive Congestion Shares (RCS), that provides bandwidth allocations independent of which congestion control algorithms flows use but consistent with the Internet's economics. We show that RCS achieves this goal using game-theoretic calculations and simulations as well as network emulation.

This paper is focused not on the Internet architecture - as defined by layering, the narrow waist of IP, and other core design principles - but on the Internet infrastructure, as embodied in the technologies and organizations that provide Internet service. In this paper we discuss both the challenges and the opportunities that make this an auspicious time to revisit how we might best structure the Internet's infrastructure. Currently, the tasks of transit-between-domains and last-mile-delivery are jointly handled by a set of ISPs who interconnect through BGP. In this paper we propose cleanly separating these two tasks. For transit, we propose the creation of a "public option" for the Internet's core backbone. This public option core, which complements rather than replaces the backbones used by large-scale ISPs, would (i) run an open market for backbone bandwidth so it could leverage links offered by third-parties, and (ii) structure its terms-of-service to enforce network neutrality so as to encourage competition and reduce the advantage of large incumbents.

A cache memory unit needs to be shared among n strategic agents. Each agent has different preferences over the files to be brought into memory. The goal is to design a mechanism that elicits these preferences in a truthful manner and outputs a fair and efficient memory allocation. A trivially truthful and fair solution would isolate each agent to a 1/n fraction of the memory. However, this could be very inefficient if the agents have similar preferences and, thus, there is room for cooperation. On the other hand, if the agents are not isolated, unless the mechanism is carefully designed, they have incentives to misreport their preferences and free ride on the files that others bring into memory. In this paper we explore the power and limitations of truthful mechanisms in this setting. We demonstrate that mechanisms blocking agents from accessing parts of the memory can achieve improved efficiency guarantees, despite the inherent inefficiencies of blocking.

In the single-resource dynamic fair division framework there is a homogeneous resource that is shared between agents dynamically arriving and departing over time. When n agents are present, there is only one truly ``fair'' allocation: each agent receives 1/n of the resource. Implementing this static solution in the dynamic world is notoriously impractical; there are too many disruptions to existing allocations: for a new agent to get her fair share, all other agents must give up a small piece.

In this paper we present an analysis of dynamic fair division of a divisible resource, with arrivals and departures of agents. Our key requirement is that we wish to disrupt the allocation of at most a small number of existing agents whenever a new agent arrives. We construct optimal recursive mechanisms to compute the allocations and provide tight analytic bounds. Our analysis relies on a linear programming formulation and a reduction of the feasible region of the LP into a class of "harmonic allocations", which play a key role in the trade-off between the fairness of current allocations and the fairness of potential future allocations. We show that there exist mechanisms that are optimal with respect to fairness and are also Pareto efficient, which is of fundamental importance in computing applications, as system designers loathe to waste resources. In addition, our mechanisms satisfy a number of other desirable game theoretic properties.

We present a model for fair strategyproof allocations in a realistic model of cloud computing centers. This model has the standard Leontief preferences but also captures a key property of virtualization, the use of containers to isolate jobs. We first present several impossibility results for deterministic mechanisms in this setting. We then construct an extension of the well known dominant resource fairness mechanism (DRF), which somewhat surprisingly does not involve the notion of a dominant resource. Our mechanism relies on the connection between the DRF mechanism and the Kalai-Smorodinsky bargaining solution; by computing a weighted max-min over the convex hull of the feasible region we can obtain an ex-ante fair, efficient and strategyproof randomized allocation. This randomized mechanism can be used to construct other mechanisms which do not rely on users' being expected (ex-ante) utility maximizers, in several ways. First, for the case of $m$ identical machines one can use the convex structure of the mechanism to get a simple mechanism which is approximately ex-post fair, efficient and strategyproof. Second, we present a more subtle construction for an arbitrary set of machines, using the Shapley-Folkman-Starr theorem to show the existence of an allocation which is approximately ex-post fair, efficient and strategyproof. This paper provides both a rigorous foundation for developing protocols that explicitly utilize the detailed structure of the modern cloud computing hardware and software, and a general method for extending the dominant resource fairness mechanism to more complex settings.

There has been a recent industrial effort to develop multi-resource hierarchical schedulers. However, the existing implementations have some shortcomings in that they might leave resources unallocated or starve certain jobs. This is because the multi-resource setting introduces new challenges for hierarchical scheduling policies. We provide an algorithm, which we implement in Hadoop, that generalizes the most commonly used multi-resource scheduler, DRF [1], to support hierarchies. Our evaluation shows that our proposed algorithm, H-DRF, avoids the starvation and resource inefficiencies of the existing open-source schedulers and outperforms slot scheduling.

In large systems, it is important for agents to learn to act effectively, but sophisticated multi-agent learning algorithms generally do not scale. An alternative approach is to find restricted classes of games where simple, efficient algorithms converge. It is shown that stage learning efficiently converges to Nash equilibria in large anonymous games if best-reply dynamics converge. Two features are identified that improve convergence. First, rather than making learning more difficult, more agents are actually beneficial in many settings. Second, providing agents with statistical information about the behavior of others can significantly reduce the number of observations needed.

Many current distributed systems users that will are satiable; users will stop providing service to others if they are themselves receiving a sufficient quantity of service. This is often the product of "tit-for-tat-like" designs, which attempt to combat free riding by denying service to those who are not providing it. While this approach provides an incentive for cooperation, it has the unfortunate side effect that if there is no service for a peer to provide, then he will generally receive reduced or no service. Ironically, this opens the systems up to an attack that we call the lotus-eater attack: the attacker supplies the service to some peers, thus satiating them. Once those peers are satiated, they stop providing service to others. The peers not being satiated by the attacker then receive reduced or no service.

We discuss the design of efficient scrip systems and develop tools for empirically analyzing them. For those interested in the empirical study of scrip systems, we demonstrate how characteristics of agents in a system can be inferred from the equilibrium distribution of money. From the perspective of a system designer, we examine the effect of the money supply on social welfare and show that social welfare is maximizedby increasing the money supply up to the point that the system experiences a "monetary crash," where money is sufficiently devalued that no agent is willing to perform a service. We alsoexamine the implications of the presence of altruists and hoarders on the performance of the system. While a small number of altruists may improve social welfare, too many can also cause the system to experience a monetary crash, which may be bad for social welfare. Hoarders generally decrease social welfare but, surprisingly, they also promote system stability by helping prevent monetary crashes. In addition, we provide new technical tools for analyzing and computing equilibria by showing that our model exhibits strategic complementarities, which implies that there exist equilibria in pure strategies that can be computed efficiently.

A model of providing service in a P2P network is analyzed. It is shown that by adding a scrip system, a mechanism that admits a reasonable Nash equilibrium that reduces free riding can be obtained. The effect of varying the total amount of money (scrip) in the system on efficiency (i.e., social welfare) is analyzed, and it is shown that by maintaining the appropriate ratio between the total amount of money and the number of agents, efficiency is maximized. The work has implications for many online systems, not only P2P networks but also a wide variety of online forums for which scrip systems are popular, but formal analyses have been lacking.

We consider the problem of designing mechanisms for online problems in which agents arrive over time and truthfully announce their arrival. These problems are becoming extremely common in a wide variety of problems involving wireless networking and webserving. We show how the standard results of mechanism design can be modified to apply to this setting, provide conditions under which efficient and incentive compatible mechanisms exist and analyze several important online models including wireless networks and web serving.

We consider the problem of designing a preemptive protocol that is both fair and efficient when one is only concerned with the sojourn time of the job and not intermediate results. Our Fair Sojourn Protocol (FSP) is both efficient, in a strong sense (similar to the shortest remaining processing time protocol: SRPT), and fair, in the sense of guaranteeing that it weakly outperforms processor sharing (PS) for every job on any sample path.Our primary motivation is web serving in which the standard protocol is PS, while recent work proposes using SRPT or variants. Our work suggests both a framework in which to evaluate proposed protocols and an attractive new protocol, FSP.