Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications最新文献_第3页

Implementation and evaluation of a parallel PMS simulator 并行PMS模拟器的实现与评价

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77166

Manohar Rao, Z. Segall

{"title":"Implementation and evaluation of a parallel PMS simulator","authors":"Manohar Rao, Z. Segall","doi":"10.1109/PARBSE.1990.77166","DOIUrl":"https://doi.org/10.1109/PARBSE.1990.77166","url":null,"abstract":"The sources of parallelism in an event-driven simulator have been investigated. First, a simple and straightforward parallelization strategy was implemented. It was observed that this strategy did not provide sufficient parallelism. Then the advance scheduling strategy was developed, which has yielded a significantly higher degree of parallelism. The main contribution of this research has been the development of this advance scheduling strategy. This strategy trades a small amount of inaccuracy in the simulation results for a large gain in the amount of parallelism extracted from the simulator. This strategy can be effectively applied for statistical simulators, because the statistical methods of simulation themselves introduce variations in the results. It was observed that, with a proper choice for the overlap distance, the error in the final results is comparable to the variation in the results introduced by the statistical simulation methods employed. A modified implementation of the event queue (Runqueue), which exploits the disorder in event scheduling created by the advance scheduling strategy, is presented. This implementation should significantly reduce contention for the Runqueue. A simulation model was implemented to simulate a distributed memory architecture proposed by D. Black, Z. Segall, and L. Rudolph. With the help of this simulation model, extensive experiments can be conducted to determine the strengths and weaknesses of the proposed architecture.<<ETX>>","PeriodicalId":389644,"journal":{"name":"Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications","volume":"68 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131859059","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A data flow architecture for parallel computation 一种用于并行计算的数据流架构

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77130

P. Abellard, G. Nolibe, N. Razafindrakoto

引用次数: 1

The multibackend database system (MDBS): a performance study 多后端数据库系统(mdb):性能研究

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77133

J. Hall, D. K. Hsiao, M. Kamel

引用次数: 2

An accelerating processor for relational operations 用于关系运算的加速处理器

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77224

H. Takeda, T. Satoh

引用次数: 2

Parallel function invocation in a dynamic argument-fetching dataflow architecture 动态参数获取数据流架构中的并行函数调用

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77126

G. Gao, H. Hum, Y. Wong

引用次数: 15

Performance analysis of a cache-coherent multiprocessor based on hierarchical multiple buses 基于分层多总线的缓存相干多处理器性能分析

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77149

Qing Yang

引用次数: 3

Dynamic alternation between receiver-initiated and sender-initiated load sharing 接收方发起和发送方发起的负载共享之间的动态交替

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77223

A. Svensson

引用次数: 2

A parallel processing strategy for computing transitive closure of a database relation 一种计算数据库关系传递闭包的并行处理策略

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77115

Keh-Chang Guh

引用次数: 3

Prolog oriented architecture using RISC concepts 使用RISC概念的面向Prolog的体系结构

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77131

Hatim Aboalsamh

引用次数: 0

Impact of large memory on the performance of optimistic concurrency control schemes 大内存对乐观并发控制方案性能的影响

Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications Pub Date : 1990-03-07 DOI: 10.1109/PARBSE.1990.77120

Philip S. Yu, D. Dias

{"title":"Impact of large memory on the performance of optimistic concurrency control schemes","authors":"Philip S. Yu, D. Dias","doi":"10.1109/PARBSE.1990.77120","DOIUrl":"https://doi.org/10.1109/PARBSE.1990.77120","url":null,"abstract":"Under optimistic concurrency control (OCC) schemes, the buffer hit ratio and hence the abort probability of a rerun transaction can be affected by its previous runs, since the data items brought in from the previous runs may still be in memory. It is noted that this buffering effect on rerun transactions has been ignored in previous performance studies. In the present work the authors examine its effect on different OCC schemes. It is shown that, with sufficient buffer, a new approach to buffer management can be adopted so that data items referenced by aborted transactions continue to be retained in memory for access during rerun. By considering the I/O reduction during rerun, it is found that, at high contention levels, the broadcast OCC which attempts to abort conflicting transactions as early as possible can be inferior to the pure OCC which only aborts a transaction at its commit time. Second, combining the two schemes, with pure OCC during the first run of a transaction and broadcast OCC during any reruns, can typically lead to better performance, especially at high contention levels.<<ETX>>","PeriodicalId":389644,"journal":{"name":"Proceedings. PARBASE-90: International Conference on Databases, Parallel Architectures, and Their Applications","volume":"210 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122409783","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 17