An atomic commit protocol for gigabit-networked distributed database systems

Al-Houmaily, Yousef J.; Chrysanthis, Panos K.

doi:10.1016/s1383-7621(99)00040-5

Cited by 18 publications

(31 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…As observed in [44], 1-delay commit protocols proposed in [45,46] assumes that all processes propose 1 before an execution starts. Jiménez-Peris et al proposed a commit service which has the same latency as 2PC but allows a process to decide twice and differently.…”

Section: Low-latency Commit Protocols With Weak Semanticsmentioning

confidence: 99%

How Fast can a Distributed Transaction Commit?

Guerraoui

Wang

2017

Proceedings of the 36th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems

View full text Add to dashboard Cite

The atomic commit problem lies at the heart of distributed database systems. The problem consists for a set of processes (database nodes) to agree on whether to commit or abort a transaction (agreement property). The commit decision can only be taken if all processes are initially willing to commit the transaction, and this decision must be taken if all processes are willing to commit and there is no failure (validity property). An atomic commit protocol is said to be non-blocking if every correct process (a database node that does not fail) eventually reaches a decision (commit or abort) even if there are failures elsewhere in the distributed database system (termination property).Surprisingly, despite the importance of the atomic commit problem, little is known about its complexity. In this paper, we present, for the first time, a systematic study on the time and message complexity of the problem. We measure complexity in the executions that are considered the most frequent in practice, i.e., failure-free, with all processes willing to commit. In other words, we measure how fast a transaction can commit. Through our systematic study, we close many open questions like the complexity of synchronous non-blocking atomic commit. We also present optimal protocols which may be of independent interest. In particular, we present an effective protocol which solves what we call indulgent atomic commit that tolerates practical distributed database systems which are synchronous "most of the time".

show abstract

Section: Low-latency Commit Protocols With Weak Semanticsmentioning

confidence: 99%

How Fast can a Distributed Transaction Commit?

Guerraoui

Wang

2017

Proceedings of the 36th ACM SIGMOD-SIGACT-SIGAI Symposium on Principles of Database Systems

View full text Add to dashboard Cite

show abstract

“…Since the latency of an ACP is a deciding factor in the performance of distributed transactions, different authors have proposed the use of One-Phase Commit protocols (1PC) that reduce both the number of forced log writes and communication steps without additional overhead. In particular, we know of at least two variations of 1PC: the Coordinator Log (CL) protocol of Stamos and Cristian [12], and the Implicit Yes-Vote (IYV) protocol of Al-Houmaily and Chrysantis [1].…”

Section: A First Glance Overviewmentioning

confidence: 99%

“…1PC protocols [1,12] assume that all the transaction operations have been acknowledged (i.e., all operations have been successfully executed till completion) before the protocol is launched. This means that the Atomicity of all the local transaction branches (i.e., local Atomicity) is already ensured at commit time.…”

Section: A Closer Look To 1pcmentioning

confidence: 99%

“…A transaction is a set of operations updating shared objects and satisfying the so-called ACID properties, namely, Atomicity, Consistency, Isolation and Durability. The Atomicity property, also called all-or-nothing property, means that either the transaction successfully executes to completion and the effects of all of its operations are recorded in the 1 This work has been partially funded by the CEC under the OpenDREAMS Esprit project n°20843 accessed objects (the transaction is said to be committed), or it fails and it has no effect at all (the transaction is aborted). Consistency means that the transaction does not violate the integrity constraints of accessed shared objects, while Isolation means that the intermediate effects of a transaction are not visible to concurrent transactions (Isolation has been formalized through the theory of serializability).…”

Section: Introductionmentioning

confidence: 99%

“…More recently, the idea of One-Phase Commit (1PC) suggested by Gray in [8] has been reconsidered, and variations of 1PC, such as Coordinator Log [12] and Implicit Yes-Vote (IYV) [1] protocols, have been proposed. 1PC further reduces the cost of atomic commit by eliminating the voting phase of 2PC: two communication steps, together with their associated forced log writes, are thus eliminated.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

One-phase commit: does it make sense?

Abdallah¹,

Guerraoui²,

Pelletier³

Proceedings 1998 International Conference on Parallel and Distributed Systems (Cat. No.98TB100250)

View full text Add to dashboard Cite

1Although widely used in distributed transactional systems, the so-called Two-Phase Commit (2PC) protocol introduces a substantial delay in transaction processing, even in the absence of failures. This has led several researchers to look for alternative commit protocols that minimize the time cost associated with coordination messages and forced log writes in 2PC. In particular, variations of a One-Phase Commit (1PC) protocol have recently been proposed. Although efficient, 1PC is however rarely considered in practice because of the strong assumptions it requires from the distributed transactional system. The aim of this paper is to better identify and understand those assumptions. Through a careful look into the intrinsic characteristics of 1PC, we dissect the assumptions underlying it and we present simple techniques that minimize them. We believe that these techniques constitute a first step towards a serious reconsideration of 1PC in the transactional world.

show abstract