three of the most important issues in exploiting network servers concern how to specify services so that service-server bindings can be changed dynamically without disturbing clients, how to make clients resilient to ...
详细信息
three of the most important issues in exploiting network servers concern how to specify services so that service-server bindings can be changed dynamically without disturbing clients, how to make clients resilient to network or server failure, and how to accommodate server protocol heterogeneity to provide a single system view to the clients. A service acquisition mechanism is presented for solving these issues. this mechanism is designed under a client/service model in which the abstraction of service is a first-class entity. the components of the mechanism are discussed.< >
An F-channel can permit as much concurrency as a nonFIFO channel and yet retain the properties of a FIFO channel that leads to simplicity of reasoning in design and proofs of the correctness of distributed algorithms....
详细信息
An F-channel can permit as much concurrency as a nonFIFO channel and yet retain the properties of a FIFO channel that leads to simplicity of reasoning in design and proofs of the correctness of distributed algorithms. this is one of the major reasons that makes it preferable to a FIFO channel and make its implementation important. this and other reasons for its superiority over a FIFO channel are discussed, and it is proposed that systems should provide implementation of F-channels. Reasons are discussed for this proposal which are based on the principle that a system should provide an interface to the users that makes algorithm design easy and that leads to efficient user efforts and system operation. Implementation of an F-channel on top of a nonFIFO channel is presented.< >
In a distributed system an algorithm used to select a distinguished node or leader in the system is known as a leader election algorithm. Leader election algorithms are examined that attempt to locate the leader at a ...
详细信息
In a distributed system an algorithm used to select a distinguished node or leader in the system is known as a leader election algorithm. Leader election algorithms are examined that attempt to locate the leader at a good node (from a performance standpoint) in the system. In the preference-based approaches examined, each node in the system uses locally available information to vote for the various candidates (potential leaders) on the basis of the performance level it would realize under each of them. the preference-based leader election algorithms proposed and examined are simple, and are shown to perform almost as well as a traditional optimization-based approach to leader election.< >
Primitives for broadcast communication that have been integrated withthe Amoeba distributed operating system are introduced. the semantics of the broadcast primitives are simple and easy to understand, but are still ...
详细信息
Primitives for broadcast communication that have been integrated withthe Amoeba distributed operating system are introduced. the semantics of the broadcast primitives are simple and easy to understand, but are still powerful. the proposed primitives, for example, guarantee global ordering of broadcast messages. the proposed primitives are also efficient: a reliable broadcast can be done in just slightly more than two messages, so the performance is comparable to a remote procedure call. In addition, the primitives are flexible; user applications can, for example, trade performance against fault-tolerance.< >
A syntax-oriented model for naming feature-based specification of services is provided. the model allows a service to evolve or reconfigure in functionality by adding and removing features and still coexist with its p...
详细信息
A syntax-oriented model for naming feature-based specification of services is provided. the model allows a service to evolve or reconfigure in functionality by adding and removing features and still coexist with its previous versions. the model's two aspects are examined. Withthis model for specifying services, name server functions may be factorized from service specific functions and implemented in a generic fashion in terms of parse and match operations and function invocations. this model can provide significant extension to such naming schemes as X.500 and the Universal Naming Protocol in supporting feature-based service interfaces.< >
Replication has been primarily used as a means of increasing availability in distributed systems. It is known that replication can mitigate the costs of accessing remotely stored data in distributed systems. Replicati...
详细信息
Replication has been primarily used as a means of increasing availability in distributed systems. It is known that replication can mitigate the costs of accessing remotely stored data in distributed systems. Replication control protocols in the literature have stopped short of addressing availability and performance concerns. these issues are addressed by contributing a classification of replicas with each class having different consistency requirements. Metareplicas keep track of up-to-date replicas for recently accessed objects and changes in data reference localities. thus they allow many transaction operations to synchronously execute at only a single (and often local) replica. Pseudoreplicas are non-permanent replicas that facilitate localized execution of transaction operations. True replicas are permanent replicas that increase the availability of operations and data. A replication control protocol is presented.< >
Techniques are proposed for computing a global virtual time (GVT), which is the minimum local virtual time of processes in Time Warp. the algorithm computes a conservative estimate of GVT using an approach which is co...
详细信息
Techniques are proposed for computing a global virtual time (GVT), which is the minimum local virtual time of processes in Time Warp. the algorithm computes a conservative estimate of GVT using an approach which is considerably simpler than previous algorithms for computing GVT. this algorithm does not require a global synchronization of processors. An inherent problem is GVT computation relates to handling messages in transit. Several alternatives are proposed for solving the transient message problem. the algorithm is suitable for distributed shared memory machines such as the BBN Butterfly and message passing machines with a variety of interconnection networks.< >
the potential of the action-oriented paradigm has been explored in the development of a specification language, DisCo, which can be characterized as both action-oriented and object-oriented. Its possibilities are intr...
详细信息
the potential of the action-oriented paradigm has been explored in the development of a specification language, DisCo, which can be characterized as both action-oriented and object-oriented. Its possibilities are introduced by contrasting them to the more familiar process-oriented approaches. Its execution model is state-based and leads to direct application of temporal logic in formal reasoning. Action-orientation allows a natural support for such forms of modularity that cut across process boundaries. At the same time, process-oriented abstractions are retained by object-orientation and the use of hierarchical state chart structures. the novel aspects of modularity are illustrated by a protocol example. the language is semi-executable.< >
A massively parallel optimization approach based on simple neighborhood search techniques was developed and applied to the problem of VLSI cell placement. Statistical models are developed to analyze the performance of...
详细信息
A massively parallel optimization approach based on simple neighborhood search techniques was developed and applied to the problem of VLSI cell placement. Statistical models are developed to analyze the performance of the approach in general, and to derive statistical bounds on the quality of obtainable results. the results of these analyses suggest a simple framework for approximate solution of difficult problems. the approach is inherently parallel, and it can be implemented on any type of parallel computer. It was implemented on a simulated hypercube MIMD machine using Cosmic C on a network of Sun workstations. the method is empirically verified.< >
Implementation of a kernel-level consistency control mechanism called invocation-based consistency control (IBCC) is presented. IBCC supports general-purpose persistent object-based distributedcomputing. It provides ...
详细信息
Implementation of a kernel-level consistency control mechanism called invocation-based consistency control (IBCC) is presented. IBCC supports general-purpose persistent object-based distributedcomputing. It provides mechanisms that support a range of powerful, well-defined consistency semantics. IBCC mechanisms can also be used to implement custom recovery and synchronization. An operating-system-level implementation of IBCC as part of the Clouds distributed operating system that uses memory faulting to initiate locking and intermediate version creation is also given. Performance aspects are discussed, as well as the overhead incurred by supporting IBCC in terms of additional data structures needed in the operating system, and the additional amount of required code.< >
暂无评论