An intelligent adaptive participant's presumption protocol for atomic commitment in distributed databases
Yousef J. Al Houmaily
Abstract
Yousef J. Al Houmaily
Abstract
The objective of the original ‘adaptive participant’s presumption protocol’ (AP3) was to achieve the performance advantages of one-phase commit protocols, whenever possible, and, at the same time, the wide applicability of two-phase commit protocols. Specifically, AP3 operates as a one-phase commit protocol for the commit processing of a transaction so long as the transaction is not associated with the validation of any deferred consistency constraints. But, when a transaction is associated with the validation of deferred consistency constraints that need to be synchronised at commit time of the transaction, AP3 dynamically switches to the most appropriate two-phase commit variant. However, providing further intelligence to enhance the performance and applicability of the original AP3 is expected to make it a highly appealing choice for adoption in the design of future distributed database management systems. This is the essence of the ‘intelligent AP3’ that extends the basic AP3 in two directions: efficiency and applicability.
OpenAlex reports 9 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The objective of the original ‘adaptive participant’s presumption protocol’ (AP3) was to achieve the performance advantages of one-phase commit protocols, whenever possible, and, at the same time, the wide applicability of two-phase commit protocols. Specifically, AP3 operates as a one-phase commit protocol for the commit processing of a transaction so long as the transaction is not associated with the validation of any deferred consistency constraints. But, when a transaction is associated with the validation of deferred consistency constraints that need to be synchronised at commit time of the transaction, AP3 dynamically switches to the most appropriate two-phase commit variant. However, providing further intelligence to enhance the performance and applicability of the original AP3 is expected to make it a highly appealing choice for adoption in the design of future distributed database management systems. This is the essence of the ‘intelligent AP3’ that extends the basic AP3 in two directions: efficiency and applicability.
Key concepts: Commit, Two-phase commit protocol, Computer science, Distributed transaction, Compensating transaction, Consistency (knowledge bases), Transaction processing, Protocol (science)