2004Unpublished venueRequires access

Analysis of cooperation semantics for transaction processing with full and partial aborts in active database

Divakar Yadav, R. K. Agrawal, R.C. Saraswat

Open publisher page 1 citations

Abstract

Active database with ECA rule has been found suitable to provide a framework to capture database events. Processing a long duration transaction in active database is a complex affair as in case of aborts, a large amount of the work has to be undone. These transactions are also more vulnerable to failures. The long duration activities can be modeled as long duration transaction where a transaction may be viewed as collection of sub transactions. Saga, a long duration transaction model, can be expressed as a series of sub transactions and provides high degree of concurrency by releasing the locks as early as possible. Transaction processing system serving cooperative and long duration transaction in active database management system exploits the semantics of transaction to achieve better processor utilization and throughput. Our simulation studies shows quantitative analysis of transaction processing with various degree of cooperation at full and partial aborts in active database.

About this research paper

What this paper is about

Active database with ECA rule has been found suitable to provide a framework to capture database events. Processing a long duration transaction in active database is a complex affair as in case of aborts, a large amount of the work has to be undone. These transactions are also more vulnerable to failures. The long duration activities can be modeled as long duration transaction where a transaction may be viewed as collection of sub transactions. Saga, a long duration transaction model, can be expressed as a series of sub transactions and provides high degree of concurrency by releasing the locks as early as possible. Transaction processing system serving cooperative and long duration transaction in active database management system exploits the semantics of transaction to achieve better processor utilization and throughput. Our simulation studies shows quantitative analysis of transaction processing with various degree of cooperation at full and partial aborts in active database.

Why it matters

OpenAlex reports 1 citations for this work. Citation counts describe recorded attention and do not establish research quality.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

Active database with ECA rule has been found suitable to provide a framework to capture database events. Processing a long duration transaction in active database is a complex affair as in case of aborts, a large amount of the work has to be undone. These transactions are also more vulnerable to failures. The long duration activities can be modeled as long duration transaction where a transaction may be viewed as collection of sub transactions. Saga, a long duration transaction model, can be expressed as a series of sub transactions and provides high degree of concurrency by releasing the locks as early as possible. Transaction processing system serving cooperative and long duration transaction in active database management system exploits the semantics of transaction to achieve better processor utilization and throughput. Our simulation studies shows quantitative analysis of transaction processing with various degree of cooperation at full and partial aborts in active database.

Key concepts: Computer science, Distributed transaction, Transaction processing system, Database transaction, Online transaction processing, Transaction processing, Duration (music), Database

Related papers

Back to paper searchBrowse research topicsOriginal source
Analysis of cooperation semantics for transaction processing with full and partial aborts in active database — Research Paper | ScholarLens