Zzyzx is a Byzantine fault-tolerant replicated state machine protocol that outperforms prior approaches and provides near-linear throughput scaling. Using a new technique called Byzantine Locking, Zzyzx allows a client to extract state from an underlying replicated state machine and access it via a second protocol specialized for use by a single client. This second protocol requires just one round-trip and 2 f + 1 responsive servers-compared to Zyzzyva, this results in 39-43% lower response times and a factor of 2.2-2.9× higher throughput. Furthermore, the extracted state can be transferred to other servers, allowing non-overlapping sets of servers to manage different state. Thus, Zzyzx allows throughput to be scaled by adding servers when concurrent data sharing is not common. When data sharing is common, performance can match that of the underlying replicated state machine protocol.

Erasure coding can reduce the space and band width overheads of redundancy in fault-tolerant data storage and delivery systems. But it introduces the fundamental difficulty of ensuring that all erasure-coded fragments correspond to the same block of data. Without such assurance, a different block may be reconstructed from different subsets of fragments. This paper develops a technique for providing this assurance without the bandwidth and computational overheads associated with current approaches. The core idea is to distribute with each fragment what we call homomorphic fingerprints. These fingerprints preserve the structure of the erasure code and allow each fragment to be independently verified as corresponding to a specific block. We demonstrate homomorphic fingerprinting functions that are secure, efficient, and compact.

This paper presents an erasure-coded Byzantine fault-tolerant block storage protocol that is nearly as efficient as protocols that tolerate only crashes. Previous Byzantine fault-tolerant block storage protocols have either relied upon replication, which is inefficient for large blocks of data when tolerating multiple faults, or a combination of additional servers, extra computation, and versioned storage. To avoid these expensive techniques, our protocol employs novel mechanisms to optimize for the common case when faults and concurrency are rare. In the common case, a write operation completes in two rounds of communication and a read completes in one round. The protocol requires a short checksum comprised of cryptographic hashes and homomorphic fingerprints. It achieves throughput within 10% of the crash-tolerant protocol for writes and reads in failure-free runs when configured to tolerate up to 6 faulty servers and any number of faulty clients.

The Domain Name System (DNS) is a critical part of the Internet's infrastructure, and is one of the few examples of a robust, highlyscalable, and operational distributed system. Although a few studies have been devoted to characterizing its properties, such as its workload and the stability of the top-level servers, many key components of DNS have not yet been examined. Based on large-scale measurements taken from servers in a large content distribution network, we present a detailed study of key characteristics of the DNS infrastructure, such as load distribution, availability, and deployment patterns of DNS servers. Our analysis includes both local DNS servers and servers in the authoritative hierarchy. We find that (1) the vast majority of users use a small fraction of deployed name servers, (2) the availability of most name servers is high, and (3) there exists a larger degree of diversity in local DNS server deployment and usage than for authoritative servers. Furthermore, we use our DNS measurements to draw conclusions about federated infrastructures in general. We evaluate and discuss the impact of federated deployment models on future systems, such as Distributed Hash Tables.