High availability in peer-to-peer DHTs requires data redundancy. This paper compares two popular redundancy schemes: replication and erasure coding. Unlike previous comparisons, we take the characteristics of the nodes that comprise the overlay into account, and conclude that in some cases the benefits from coding are limited, and may not be worth its disadvantages.

Geo-replicated storage systems are at the core of current Internet services. The designers of the replication protocols used by these systems must choose between either supporting low-latency, eventually-consistent operations, or ensuring strong consistency to ease application correctness. We propose an alternative consistency model, Explicit Consistency, that strengthens eventual consistency with a guarantee to preserve specific invariants defined by the applications. Given these application-specific invariants, a system that supports Explicit Consistency identifies which operations would be unsafe under concurrent execution, and allows programmers to select either violation-avoidance or invariant-repair techniques. We show how to achieve the former, while allowing operations to complete locally in the common case, by relying on a reservation system that moves coordination off the critical path of operation execution. The latter, in turn, allows operations to execute without restriction, and restore invariants by applying a repair operation to the database state. We present the design and evaluation of Indigo, a middleware that provides Explicit Consistency on top of a causally-consistent data store. Indigo guarantees strong application invariants while providing similar latency to an eventually-consistent system in the common case.

Sofhyare errors are a major cause of outages and they are increasingly exploited in malicious attacks. Byzantine fault tolerance allows replicated systems to mask some software errors but it is expensive to deploy. This paper describes a replication technique, BFTA, which uses abstraction to reduce the cost of Byzantine fault tolerance and to improve its ability to mask software errors. BFTA reduces cost because it enables reuse of off-the-shelf service implementations. It improves availability because each replica can be repaired periodically using an abstract view of the state stored by correct replicas, and because each replica can run distinct or non-deterministic service implementations, which reduces the probability of common mode failures. We built an NFS service that allows each replica to run a different operating system. This example suggests that BFTA can be used in practice -the replicatedjile system required only a modest amount of new code, and preliminary pe~ormance results indicate that it pe$orms comparably to the off-the-shelf implementations that it wraps.

reviewarticles Peer-tO-Peer (P 2 P ) cO M P Ut i nG has attracted significant interest in recent years, originally sparked by the release of three influential systems in 1999: the Napster music-sharing system, the Freenet anonymous data store, and the SETI@home volunteerbased scientific computing projects. Napster, for instance, allowed its users to download music directly from each other's computers via the Internet. Because the bandwidth-intensive music downloads occurred directly between users' computers, Napster avoided significant operating costs and was able to offer its service to millions of users for free. Though unresolved legal issues ultimately sealed Napster's fate, the idea of cooperative resource sharing among peers found its way into many other applications.More than a decade later, P2P technology has gone far beyond music sharing, anonymous data storage, or scientific computing; it now enjoys significant research attention and increasingly widespread use in open software communities and industry alike. Scientists, companies, and open-software Peer-to-Peer Systems key insights P2P leverages the computing resources of cooperating users to achieve scalability and organic growth, thus lowering the deployment barrier for innovative new services. originally invented for music/data sharing and volunteer computing, P2P systems now enjoy widespread commercial and non-commercial use in content distribution, iPtV, and iP telephony.the strength of P2P-its independence of dedicated infrastructure and centralized control-is also its weakness, as it presents new technical, commercial, and legal challenges.P2P technology may turn out to be most valuable as a low-cost deployment vector for experimental, innovative services; those services that prove to be commercially viable can be subsequently combined with centralized, infrastructure-based components. iLLustratio n by Ma rius wat Z cred it t K O c TO b e r 2 0 1 0 | VO L. 5 3 | n O. 1 0 | c o m m u n i c at i o nS o f t he acm 73

We propose a specification for weak consistency in the context of a replicated service that tolerates Byzantine faults. We define different levels of consistency for the replies that can be obtained from such a service-we use a real world application that can currently only tolerate crash faults to exemplify the need for such consistency guarantees.

Abstract

No abstract

Most peer-to-peer lookup schemes keep a small amount of routing state per node, typically logarithmic in the number of overlay nodes. This design assumes that routing information at each member node must be kept small so that the bookkeeping required to respond to system membership changes is also small, given that aggressive membership dynamics are expected. As a consequence, lookups have high latency as each lookup requires contacting several nodes in sequence. In this paper, we question these assumptions by presenting a peer-to-peer routing algorithm with small lookup paths. Our algorithm, called "OneHop," maintains full information about the system membership at each node, routing in a single hop whenever that information is up to date and in a small number of hops otherwise. We show how to disseminate information about membership changes quickly enough so that nodes maintain accurate complete membership information. We also present analytic bandwidth requirements for our scheme that demonstrate that it could be deployed in systems with hundreds of thousands of nodes and high churn. We validate our analytic model using a simulated environment and a real implementation. Our results confirm that OneHop is able to achieve high efficiency, usually reaching the correct node directly 99 percent of the time.