Very large databasesystems require distributed storage, which means that they need distributed search structures for fast and efficient access to the data. In this paper, we present an approach to maintaining distrib...
详细信息
ISBN:
(纸本)0897915925
Very large databasesystems require distributed storage, which means that they need distributed search structures for fast and efficient access to the data. In this paper, we present an approach to maintaining distributed data structures that uses lazy updates, which take advantage of the semantics of the search structure operations to allow for scalable and low-overhead replication. Lazy updates can be used to design distributed search structures that support very high levels of concurrency. The alternatives to lazy update algorithms (eager updates) use synchronization to ensure consistency, while lazy update algorithms avoid blocking. Since lazy updates avoid the use of synchronization, they are much easier to implement than eager update algorithms. We demonstrate the application of lazy updates to the dB-tree, which is distributed B+ tree that replicates its interior nodes for highly parallel access. We develop a correctness theory for lazy updates so that our algorithms can be applied to other distributed search structures.
The advent of client/server computing has forced the old character-mode 4GL vendors to respond with new products. The leading vendors have reengineered their products with stored procedures and triggers to support an ...
详细信息
The advent of client/server computing has forced the old character-mode 4GL vendors to respond with new products. The leading vendors have reengineered their products with stored procedures and triggers to support an event-driven model, cross-platform graphic user interfaces and open database access.
Existing and legacy software systems are the product of lengthy and individual developmental histories. Interoperability among such systems offers the support of global applications on these systems and intelligent in...
详细信息
Existing and legacy software systems are the product of lengthy and individual developmental histories. Interoperability among such systems offers the support of global applications on these systems and intelligent information processing. However, interoperability among these heterogeneous systems is hampered by the absence of an integrated environment that would allow the development of global applications requiring intersystem cooperation. A uniform application-system interface is necessary to abstract the common properties of the global applications and of systems, mask their differences, and thus overcome this heterogeneity barrier. This paper presents such a solution, termed Remote System Interfaces (RSIs), which has been designed and implemented in the course of the InterBase project at Purdue University.
Alsys has implemented a distributed Ada system known as AdaMap which is based on a program partitioning model almost identical to that being proposed for Ada9X, although the user interface for the inter-partition comm...
详细信息
Alsys has implemented a distributed Ada system known as AdaMap which is based on a program partitioning model almost identical to that being proposed for Ada9X, although the user interface for the inter-partition communication system is based on an input-output package rather than on remote procedure call. This paper examines the benefits of the partitions model including a simple demonstration program built using AdaMap, and is intended to give confidence in the viability of the partitions model for Ada9X.
The TPC-C benchmark is a new benchmark approved by the TPC council intended for comparing database platforms running a medium complexity transaction processing workload. Some key aspects in which this new benchmark di...
详细信息
VODAK is a prototype of an object-oriented, distributeddatabase system developed during the past five years at the Integrated Publication and Information systems Institute (IPSI). The aim of demonstrating VODAK Open ...
详细信息
This paper introduces a simple and efficient distributed deadlock detection algorithm for a distributeddatabase System with AND model. The algorithm is derived from a previous algorithm for the One-Resource model. By...
详细信息
This paper introduces a simple and efficient distributed deadlock detection algorithm for a distributeddatabase System with AND model. The algorithm is derived from a previous algorithm for the One-Resource model. By using a formal specification based on Automata Theory, the proof of correctness is provided: the algorithm reports all true deadlocks and does not report false deadlocks.
Each transaction in a real-time database system is associated with a timing constraint in the form of a deadline. Efficient transaction scheduling algorithms are required to minimize the number of missed transaction d...
详细信息
ISBN:
(纸本)0818627654
Each transaction in a real-time database system is associated with a timing constraint in the form of a deadline. Efficient transaction scheduling algorithms are required to minimize the number of missed transaction deadlines. In this paper, a performance evaluation model is provided to enable distributed real-time database system designers to analyze transaction scheduling algorithms. The model is developed progressing from a simple mathematical analysis to complicated simulations. The paper also provides an example simulation experiment implemented using the model presented.
Data distribution, data replication and system reliability are key factors in determining the availability measures for transactions in distributed database systems. In order to simplify the evaluation of these measur...
详细信息
Data distribution, data replication and system reliability are key factors in determining the availability measures for transactions in distributed database systems. In order to simplify the evaluation of these measures, database designers and researchers tend to make unrealistic assumptions about these factors. In this article, we investigate the effect of such assumptions on the computational complexity and accuracy of such evaluations. We represent a database system with five parameters related to the above factors. Probabilistic analysis is employed to evaluate the availability of read-only and read-write transactions. We consider both the read-one/write-all and the majority-read/majority-write replication control policies. We conclude that transaction availability is more sensitive to variations in degrees of replication, less sensitive to data distribution, and insensitive to reliability variations in a nonhomogeneous system. The computational complexity of the evaluations is found to be mainly determined by the chosen distributeddatabase model, while the accuracy of the results is not so much dependent on the models.
A queueing model is used to approximate static locking in distributed database systems without deadlocks. Then a random graph model to find the deadlock probability of each transaction is proposed. Finally, the above ...
详细信息
A queueing model is used to approximate static locking in distributed database systems without deadlocks. Then a random graph model to find the deadlock probability of each transaction is proposed. Finally, the above two models are integrated, so that given the transaction arrival rate, the response time and the effective throughput can be calculated.
暂无评论