To provide high availability for services such as mail or bulletin boards, data must be replicated. One way to guarantee consistency of replicated data is to force service operations to occur in the same order at all sites, but this approach is expensive. For some applications a weaker causal operation order can preserve consistency while providing better performance. This paper describes a new way of implementing causal operations. Our technique also supports two other kinds of operations: operations that are totally ordered with respect to one another, and operations that are totally ordered with respect to all other operations. The method performs well in terms of response time, operation processing capacity, amount of stored state, and number and size of messages; it does better than replication methods based on reliable multicast techniques. This research was supported in part by the National Science Foundation under Grant CCR-8822158 and in part by the Advanced Research Projects ...

To provide high availability for services such as mail or bulletin boards, data must be replicated. One way to guarantee consistency of replicated data is to force service operations to occur in the same order at all sites, but this approach is expensive. In this paper, we propose lazy replication as a way to preserve consistency by exploiting the semantics of the service's operations to relax the constraints on ordering. Three kinds of operations are supported: operations for which the clients define the required order dynamically during the execution, operations for which the service defines the order, and operations that must be globally ordered with respect to both client ordered and service ordered operations. The method performs well in terms of response time, amount of stored state, number of messages, and availability. It is especially well suited to applications in which most operations require only the client-defined order.

The dissertation presents the issues addressed in the design of Ficus, a large scale wide area distributed file system currently operational on a modest scale at UCLA. Key aspects of providing such a service include toleration of partial operation in virtually all areas; support for large scale, optimistic data replication; and a flexible, extensible modular design. Ficus incorporates a "stackable layers" modular architecture and full support for optimistic replication. Replication is provided by a pair of layers operating in concert above a traditional filing service. A "volume" abstraction and on-the-fly volume "grafting" mechanism are used to manage the large scale file name space. The replication service uses a f...

. Many distributed databases use an epidemic approach to manage replicated data. In this approach, user operations are executed on a single replica. Asynchronously, a separate activity performs periodic pair-wise comparison of data item copies to detect and bring up to date obsolete copies. The overhead due to comparison of data copies grows linearly with the number of data items in the database, which limits the scalability of the system. We propose an epidemic protocol whose overhead is linear in the number of data items being copied during update propagation. Since this number is typically much smaller than the total number of data items in the database, our protocol promises significant reduction of overhead. 1 Introduction Data replication is often used in distributed systems to improve system availability and performance. Examples of replicated systems abound and include both research prototypes (e.g., [5, 14]) and commercial systems (e.g., [8, 10]). Many of these systems use an...

The need for replication of information storage, with system support for maintaining consistency among the various copies, has been widely recognized. However, earlier work guaranteed that copies would be consistent by assuring that conflicting updates could not occur. Such an approach decreases availability for update, often at precisely the point when availability is desired. Further, there is increasing data that conflicting updates are extremely rare in many situations of interest. This paper presents a practical set of algorithms for maintaining the consistency of a replicated file system with an optimistic update policy. These algorithms permit a system which allows updates to an object so long as any copy is available; the algorithms then return the various copies to consistency at their first opportunity. These algorithms have been used to build the Ficus replicated file system. 1 Introduction Replication of important information can serve to improve both the performance and f...

We present a family of epidemic algorithms for maintaining replicated database systems. The algorithms are based on the causal delivery of log records where each record corresponds to one transaction instead of one operation. The first algorithm in this family is a pessimistic protocol that ensures serializability and guarantees strict executions. Since we expect the epidemic algorithms to be used in environments with low probability of conflicts among transactions, we develop a variant of the pessimistic algorithm which is optimistic in that transactions commit as soon as they terminate locally and inconsistencies are detected asynchronously as the effects of committed transactions propagate through the system. The last member of the family of epidemic algorithms is pessimistic and uses voting with quorums to resolve conflicts and improve transaction response time. A simulation study evaluates the performance of the protocols.

This paper presents a simple and efficient algorithm to record a global snapshot of a distributed system, where all messages are sent and delivered in causal order. For a system with N processes, the algorithm requires N control messages to record a distributed snapshot of the system. It is shown that the snapshot recorded represents a consistent global state of the system.

. Update propagation and transaction atomicity are major obstacles to the development of replicated databases. Many practical applications, such as automated teller machine networks, flight reservation, and part inventory control, do not require these properties. In this paper we present an approach for incrementally updating a distributed, replicated database without requiring multi-site atomic commit protocols. We prove that the mechanism is correct, as it asymptotically performs all the updates on all the copies. Our approach has two important characteristics: it is progressive, and non-blocking. Progressive means that the transaction's coordinator always commits, possibly together with a group of other sites. The update is later propagated asynchronously to the remaining sites. Non-blocking means that each site can take unilateral decisions at each step of the algorithm. Sites which cannot commit updates are brought to the same final state by means of a reconciliation mechanism. Th...

Matrix clocks have nice properties that can be used in the context of distributed database protocols and fault tolerant protocols. Unfortunately, they are costly to implement, requiring storage and communication overhead of size O(n²) for a system of n sites. They are often considered a non feasible approach when the number of sites is large. In this paper, we firstly describe an efficient incremental algorithm to compute the matrix clock, which achieves storage and communication overhead of size O(n) when the sites of the computation are "well synchronized". Secondly, we introduce the k-matrix clock: an approximation to the genuine matrix clock that can be computed with a storage and communication overhead of size O(kn). k-matrix clocks can be useful to implement fault-tolerant protocols for systems with crash failure semantics such that the maximum number of simultaneous faults is bounded by k - 1.

Abstract not found