ML-1-2PC: An Adaptive Multi-level Atomic Commit Protocol

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

doi:10.1007/978-3-540-30204-9_19

Cited by 8 publications

(9 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…where 𝑡 sent , 𝑡 recv are respectively the clock time of 𝐶 * and 𝐶 𝑖 when sending and receiving the Propose message 4 .…”

Section: Flac Ff For Failure-free Environmentmentioning

confidence: 99%

“…Algorithm 5 formalizes how the RLSM manager detects unexpected failures or consecutive failure-free executions and sends the corresponding events (Definition 2) to the participants' RLSMs. Firstly, if the RLSM manager can detect a crash or a network failure during a transaction execution (lines 2, 5, 8, 10), it will trigger a 𝐶𝐹 or 𝑁 𝐹 event to update the current state to a more stringent robustness level 𝐿 CF or 𝐿 NF (lines 3,4,6,9,11). This update is called the robustness upgrade transition; it corresponds to the solid lines in Figure 5.…”

Section: Robustness-level State Machinesmentioning

confidence: 99%

“…ADTP [61] can let service providers choose between PrC and PrA. ML1-2PC [4,5] can dynamically select either 1PC or 2PC based on transaction behavior and system requirements. Besides, a consensus protocol Domino [57] can adapt executions automatically between fast or slow paths with the network timeout.…”

Section: Related Workmentioning

confidence: 99%

“…To alleviate the above problems, we design FLAC, a failure-aware atomic commit protocol, which can monitor and adapt its execution to failures happening in the operating environments to improve its performance. Similar to existing ACPs [4,7,28,33,42], our system model consists of a coordinator node and different participant nodes. The coordinator drives the protocol while participants vote on the transaction status.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

FLAC: Practical Failure-Aware Atomic Commit Protocol for Distributed Transactions

Pan¹,

Ta²,

Zhang³

et al. 2023

Preprint

View full text Add to dashboard Cite

In distributed transaction processing, atomic commit protocol (ACP) is used to ensure database consistency. With the use of commodity compute nodes and networks, failures such as system crashes and network partitioning are common. It is therefore important for ACP to dynamically adapt to the operating condition for efficiency while ensuring the consistency of the database. Existing ACPs often assume stable operating conditions, hence, they are either nongeneralizable to different environments or slow in practice.In this paper, we propose a novel and practical ACP, called Failure-Aware Atomic Commit (FLAC). In essence, FLAC includes three protocols, which are specifically designed for three different environments: (i) no failure occurs, (ii) participant nodes might crash but there is no delayed connection, or (iii) both crashed nodes and delayed connection can occur. It models these environments as the failure-free, crash-failure, and network-failure robustness levels. During its operation, FLAC can monitor if any failure occurs and dynamically switch to operate the most suitable protocol, using a robustness level state machine, whose parameters are finetuned by reinforcement learning. Consequently, it improves both the response time and throughput, and effectively handles nodes distributed across the Internet where crash and network failures might occur. We implement FLAC in a distributed transactional keyvalue storage system based on Google Percolator and evaluate its performance with both a micro benchmark and a macro benchmark of real workload. The results show that FLAC achieves up to 2.22x throughput improvement and 2.82x latency speedup, compared to existing ACPs for high-contention workloads.

show abstract

“…where 𝑡 sent , 𝑡 recv are respectively the clock time of 𝐶 * and 𝐶 𝑖 when sending and receiving the Propose message 4 .…”

Section: Flac Ff For Failure-free Environmentmentioning

confidence: 99%

Section: Robustness-level State Machinesmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

FLAC: Practical Failure-Aware Atomic Commit Protocol for Distributed Transactions

Pan¹,

Ta²,

Zhang³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…This is due to the costs associated with its message complexity (i.e., the number of messages used for coordinating the actions of the different sites) and log complexity (i.e., the frequency at which information is stored onto the stable logs of the participating sites). For this reason, there is a continuous interest in developing more efficient ACPs and optimizations (e.g., [10,7,4,2,13,5,6]), especially in the context of modern electronic services and electronic commerce environments that are characterized by high volume of transactions. Most notable developments are one-phase commit (IPC) protocols [14,4,2].…”

Section: Introductionmentioning

confidence: 99%

On interoperating incompatible atomic commit protocols in distributed databases

Al-Houmaily¹

2005

2005 1st International Conference on Computers, Communications, &Amp; Signal Processing With Special Track on Biomedical Engine

Self Cite

View full text Add to dashboard Cite

This paper proposes an adaptive participant's presumption protocol (Ap3) that can be used to atomically commit Internet transactions. Ap3 interoperates a one-phase commit protocol, namely, implicit-yes vote, and two-phase commit protocols, namely, presumed abort and presumed commit, in a dynamic fashion. Thus, it offers the performance advantage of the combined protocols, whenever possible, while still providing the wide applicability of two-phase commit protocols. This is achieved in spite ofthe incompatibilities between atomic commit protocols that lead to the general practice ofadopting a single atomic commit protocol in any distributed database system.

show abstract