Ming T. Liu

Synthesizing protocol specifications from service specifications in FSM model

The authors address the problem of deriving a protocol specification from a given service specification, which are both in the finite-state-machine (FSM) model. They provide an algorithm that yields protocol FSMs from a given service FSM in which the service model allows only sequential execution of service primitives at various service access points (SAPs). They then extend the service model to allow concurrent execution of service primitives at different SAPs, and the algorithm is also expanded to take care of the extended model. Finally, they present a procedure to construct an error-recoverable protocol from its error-free version generated from the proposed algorithm, thus enabling the algorithm to provide the generated protocol with error-recovery capability. Limitations of the model and the direction being taken in further work to relieve the limitations are stated.<<ETX>>

An Analysis of the Effect of Network Parameters on the Performance of Distributed Database Systems

Performance analysis studies of distributed database systems in the past have assumed that the message transmission time between any two nodes of a network is constant. They disregard the effect of communication network parameters such as network traffic, network topology, and capacity of transmission channels. In this paper, an analytical model is used to estimate the delays in transmission channels of the long haul network supporting the distributed database system. The analysis shows that the constant transmission time assumption cannot be justified in many cases, and that the response time is sensitive to the parameters mentioned above. Extensions and performance analysis in the context of interconnection networks are also discussed.

A distributed deadlock detection and resolution algorithm and its correctness proof

The key idea of the algorithm is to let one transaction controller be in charge of all transactions in a set of interacting transactions. Two transactions are interacting if they are both interested in (accessing) the same resource. In addition, the controller is in charge of all the resources allocated to any of the transactions in the set. Having one controller in charge of all the transactions in a set of interacting transactions and all the resources allocated to them makes it easier to detect deadlocks and avoid them. The main problem dealt with is how a controller takes charge of another transaction when the transaction tries to access one of the resources currently in the control of the controller and how a controller releases a transaction back to its original controller when the transaction is no longer interested in any of the resources in which one or more of the other transactions are also interested. Communicating sequential processes (CSP) is used to code the algorithm. The correctness of the algorithm is proved in a semiformal manner. >

Priority communication schemes on local area networks for multimedia traffic

We address the issues related to the delivery of multimedia streams on local area networks. Multimedia integrates voice and video along with text and images into existing systems. Traditionally, local area networks were designed to handle regular data traffic which are bursty in nature and for which variable delay is acceptable. Multimedia traffic, on the other hand, requires constant delay in addition to fast (or real time) delivery. Multimedia traffic also requires large bandwidth compared to regular traffic. Since local area networks are widely in use, their modification to integrate multimedia streams is very important. We propose priority-based communication schemes for the timely delivery of multimedia traffic on local area networks. We compare the performance of these schemes using simulation techniques.<<ETX>>

Distributed rule processing in active databases

Processing rules in a distributed active database involves three design issues: how to decompose rules, how to distribute rules to sites, and how to evaluate distributed rules correctly. The authors study these three issues for complicated rules, which are complex and time-consuming to evaluate. They propose a relational operator, AND, and the associated algebraic manipulations of this operator to find independent parts of a rule query, which can be distributed among sites. Due to geographical dispersion in a distributed system, correct evaluation of distributed rules is not trivial. A distributed evaluation algorithm is preferred, which guarantees the correctness of the evaluation result of the distributed rule by collecting consistent local results from sites to form a global view.<<ETX>>

Performance study of buffering within switches in local area networks

Todays demanding applications, such as multimedia, require higher network transfer rates. Traditional local area networks (LANs) will not be able to provide the throughput required by these applications. The use of switches in LANs is an effective technique to increase the throughput of the network. These switches have finite buffers at their input and output ports. The size of these buffers affect the packet loss rate. It also affect the delay at the switch as packets may have to wait for the output buffer to become available. We study the effect of buffer sizes within these switches. We show how the buffer size is related to the performance of the switch as well as overall performance of the LAN.

Performance evaluation of switching in local area networks

In this paper we study the performance of switching in local area networks (LAN). The transfer rates required by todays applications, such as multimedia, are beyond the typical bandwidth of conventional LANs. The use of switches in LANs overcomes the bandwidth limitation by supporting multiple concurrent communications. The overall performance including throughput and network reliability are improved. We study the performance of different switch architectures using simulation and analytical techniques. We show that parallel switches work very well with high traffic load and at light-to-medium load they give close to optimal performance. They provide much better overall performance than traditional switches.

Incremental protocol verification using deductive database systems

A deductive approach which incorporates first-order logic into a relational database is proposed to remedy the problems incurred in the relational algebraic approach. The method not only allows recursive definitions for more complex logical reasoning but also provides a uniform query interface to the users for verifying functional properties of a protocol. Most important, with the deductibility of the deductive database it becomes possible to define algorithms for the incremental verification method, which speeds up the verification process by reverifying protocols without generating the global states from scratch.<<ETX>>

An optimistic concurrency control mechanism without freezing for distributed database systems

Optimistic concurrency control mechanisms have the tendency of freezing the system in order to validate a transaction at the end of the transactions execution phase. This paper presents an optimistic concurrency control that does not freeze the database system. According to the property of conflicting transactions, concurrency control is divided into two counterparts. While one part is exercised at the data object, which responds instantly to consistency violation, the other part is exercised by messages flowing between transactions as background tasks. The whole system need not stop during the time concurrency control is exercising, nor during the time the system is restoring its state when consistency is violated. This makes the mechanism more attractive, since a higher degree of parallelism is provided between consistency control and transaction execution. An object-based model is used to present the mechanism in a distributed database system, in which communication relies heavily on message passing.

Performance analysis of resiliency mechanisms in distributed database systems

Degradation in system performance due to component failures is an important factor that prevents a distributed database management system (DDBMS) from achieving its full potential of better availability, response time, and system throughput. Resiliency mechanisms that help to continue system operation in spite of failures introduce overhead due to the need to maintain redundant information. Earlier performance studies of DDBMS have either altogether ignored failure or studied only limited aspects of the effect of failures and performance of resiliency mechanisms. In this paper an analytical model is used to comprehensively characterize the effect of failures and resiliency mechanisms on the performance of DDBMS. Two new performance measures are introduced. The methodology is illustrated by comparatively evaluating the performance of three algorithms.