Write Markers for Probabilistic Quorum Systems
نویسندگان
چکیده
Probabilistic quorum systems can tolerate a larger fraction of faults than can traditional (strict) quorum systems, while guaranteeing consistency with an arbitrarily high probability for a system with enough replicas. However, they are hampered in that, like strict quorum systems, they allow for Byzantine-faulty servers to collude maximally to provide incorrect values to clients. We present a technique based on write markers that prevents faulty servers from colluding unless they are all also selected to be participants in the same update operations. We show that write markers increase the maximum fraction of faults that can be tolerated to b < n/2 from b < n/2.62, where n is the total number of replicas, for probabilistic masking quorum systems (compared with b < n/4 for strict masking quorum systems) and to b < n/2.62 from b < n/3.15 for probabilistic opaque quorum systems (compared with b < n/5 for strict opaque quorum systems). In addition, with write markers, probabilistic masking quorums no longer require write quorums of large or maximal size in order to tolerate the maximum fraction of faults. We describe an implementation of write markers that is effective even if Byzantine clients collude with faulty servers. This work was partially supported by NSF grant CCF-0424422.
منابع مشابه
Probabilistic quorum systems for dependable distributed data management
Among failure-prone and dynamic distributed systems there is a significant class of systems that strive for high availability and can function with inconsistent data. Examples include flight reservation systems which allow overbooking or emergency ambulance systems which return informative responses to time-critical queries. Data replication is a well-known technique for tolerating failures and...
متن کاملQuorum Systems Access Protocol
Quorum systems are tools for increasing the availability and eeciency of replicated services. A quorum system for a universe of servers is a collection of subsets of servers, each pair of which intersect. Intuitively, each quorum can operate on behalf of the system, thus increasing its availability and performance, while the intersection property guarantees that operations done on distinct quor...
متن کاملMinimizing Content Staleness in Dynamo-Style Replicated Storage Systems
Consistency in data storage systems requires any read operation to return the most recent written version of the content. In replicated storage systems, consistency comes at the price of delay due to large-scale write and read operations. Many applications with low latency requirements tolerate data staleness in order to provide high availability and low operation latency. Using age of informat...
متن کاملProbabilistically Bounded Staleness for Practical Partial Quorums
Modern storage systems employing quorum replication are often configured to use partial, non-strict quorums. These systems wait only for a subset of their replicas to respond to a request before returning an answer, without guaranteeing that read and write replica sets intersect. While these partial quorum mechanisms provide only basic eventual consistency guarantees, with no limit to the recen...
متن کاملProbabilistic Quorum-Based Accounting for Peer-to-Peer Systems
Providing an effective accounting scheme for decentralized peer-to-peer systems is difficult without relying on a central server or peers with special trusted status. In this paper, we present Probabilistic Quorum-based Accounting (PQA) for peer-to-peer systems. In our approach, a peer receiving a request will retrieve resource usage accounting information about the requesting peer from a quoru...
متن کامل