One-to-many (group) interprocess communication is useful in many real-time distributed applications. It may be conveniently and efficiently realized using the multicast feature available in contemporary local area net...
详细信息
ISBN:
(纸本)0818607491
One-to-many (group) interprocess communication is useful in many real-time distributed applications. It may be conveniently and efficiently realized using the multicast feature available in contemporary local area networks. A kernel model which supports reliable group communication in a distributed computing environment is presented. New semantic tools which capture the nondeterminism of the underlying low-level events concisely are introduced and a process alias-based structuring technique for the kernel to handle the reliability problems that may arise during group communication is described. The scheme works by maintaining a close association between group messages and their corresponding reply messages. Sample programs illustrate how the semantic tools may be used.
A simple real-time load-sharing algorithm is presented in which the decision to execute a job locally or remotely is made dynamically, on the basis of a simple threshold policy. The selection of destination node at wh...
详细信息
ISBN:
(纸本)0818607491
A simple real-time load-sharing algorithm is presented in which the decision to execute a job locally or remotely is made dynamically, on the basis of a simple threshold policy. The selection of destination node at which the job is to be executed is made probabilistically and independently of the current system state. An approximate analytic performance model is developed and validated through simulation. The performance results suggest that, over a relatively wide range of system parameters, the performance of the algorithm is substantially better than that of extremely simple algorithms and often close to that of a theoretically optimum algorithm.
For systems.containing a large number of processing elements (PEs), the capability to recover from a PE fault is important. The dynamic redundancy (DR) network can tolerate faults in the network and support a system t...
详细信息
ISBN:
(纸本)0818607491
For systems.containing a large number of processing elements (PEs), the capability to recover from a PE fault is important. The dynamic redundancy (DR) network can tolerate faults in the network and support a system that tolerates PE faults without degradation in performance by adding spare PEs, while retaining the full capability of a multistage cube. A variation of the DR network, the reduced DR (RDR) network, is presented which can be implemented more cost effectively while retaining most of the advantages of the DR. The reliability of systems.containing the DR or RDR networks and spare PEs and the reliability of systems.with no spare PEs are also estimated and compared. It is shown that using the DR or RDR network and spare PEs in a system can achieve better system reliability over a wide range of N, where N is the number of functioning PEs, than can using any kind of N multiplied by N fault-tolerant network and no spares.
The authors present and analyze a receiver-initiated scheduling algorithm for distributed soft real-time systems. The algorithm is based on a 'poll when underloaded' approach. Using simulation, the deadline mi...
详细信息
ISBN:
(纸本)0818607491
The authors present and analyze a receiver-initiated scheduling algorithm for distributed soft real-time systems. The algorithm is based on a 'poll when underloaded' approach. Using simulation, the deadline miss ratio and mean lateness of the algorithm are derived for different workloads and overheads. The performance profile of the algorithm shows stable behavior when load or management overheads are high. Its performance is compared with that of sender-initiated scheduling algorithm that uses a 'first-fit' strategy. The sender-initiated algorithm outperforms the receiver-initiated algorithm when the load is light. The results show that although there is a clear need for dynamic deadline-oriented scheduling at medium to high loads, a simple algorithm would suffice at low loads.
To meet the growing demand for online transaction processing, several DB (database management) and DC (data communication management) subsystems.can be coupled together to form a distributed DB/DC system. A recovery p...
详细信息
ISBN:
(纸本)0818607491
To meet the growing demand for online transaction processing, several DB (database management) and DC (data communication management) subsystems.can be coupled together to form a distributed DB/DC system. A recovery protocol is needed not only to provide for the recovery of transactions affected by the failure, but also to localize recovery operations. Two such protocols based on a progressive approach, namely, a synchronous progressive and an asynchronous progressive protocol, along with a pessimistic protocol, are analyzed. Their performance during normal transaction processing is contrasted with that of a transaction processing system without any recovery protocol. A queuing model is developed and simulated to predict the transaction response time. The progressive recovery approach is found to reduce normal processing overhead and lead to performance improvement over the pessimistic approach.
This paper describes how new computer technology is being applied to implement distributed process control systems.with open architectures. Four design aspects are reviewed: system architecture, integrated software en...
详细信息
ISBN:
(纸本)087664891X
This paper describes how new computer technology is being applied to implement distributed process control systems.with open architectures. Four design aspects are reviewed: system architecture, integrated software environment, communication network and operational performance. Architecturally, the distribution of host computer functions and communication among processors is discussed. From a language perspective, the trend toward interpretive and table-driven software is reviewed. The integration of control subsystems.from diverse vendors through the use of global database and factory oriented local area networks is analyzed. The effects of distribution on reliability, maintainability and response time are considered.
Efforts at achieving reliability in distributed computing include the incorporation of fault-tolerance into system software such as routing tables, and the introduction of extra software to mask processor failures, su...
详细信息
ISBN:
(纸本)0818605642
Efforts at achieving reliability in distributed computing include the incorporation of fault-tolerance into system software such as routing tables, and the introduction of extra software to mask processor failures, such as Byzantine agreement. A set of n processors running software that can tolerate k faults is called a (k,n)-resilient system. Most (k,n)-resilient systems.have a higher probability of failure in the long run, and a smaller mean time-to-failure, than one processor. Further, large systems.fail in a very well-defined period, and they fail very quickly if n is more than linear in k.
An algorithm for dynamic planning of recovery line is specified. A computational model is defined for a distributed system of communicating processes using asynchronous message passing, and the recovery algorithms are...
详细信息
ISBN:
(纸本)0818605642
An algorithm for dynamic planning of recovery line is specified. A computational model is defined for a distributed system of communicating processes using asynchronous message passing, and the recovery algorithms are described by means of axioms.
An approach to the development of fault-tolerant distributedsoftware by a high-level concurrent language is discussed. The language constructs support mutual control and consensus about the decisions on the distribut...
详细信息
ISBN:
(纸本)0818605642
An approach to the development of fault-tolerant distributedsoftware by a high-level concurrent language is discussed. The language constructs support mutual control and consensus about the decisions on the distributed system state. Emphasis is placed on process structuring, parallel activation, and termination control. It is shown that both forward and backward recovery can be expressed in the proposed language. Backward recovery is based on nested atomic actions.
暂无评论