TxnSails: Achieving Serializable Transaction Scheduling with Self-Adaptive Isolation Level Selection

2025-02-03

AI Summary

To address the prohibitively high performance overhead of serializable isolation, this paper proposes a dynamic, adaptive transaction scheduling framework. The method introduces: (1) a unified concurrency control algorithm enabling co-execution across multiple isolation levels; (2) a deep learning–driven, workload-aware model that predicts and selects the lowest feasible isolation level guaranteeing serializability in real time; and (3) a cross-isolation-level consistency verification mechanism ensuring semantic correctness at runtime. Experimental evaluation demonstrates that, while strictly preserving serializability, the approach achieves up to 26.7× higher throughput than state-of-the-art (SOTA) serializable systems and 4.8× higher throughput than PostgreSQL’s serializable mode. These results significantly alleviate the inherent tension between strong consistency and high performance in transactional database systems.

Technology Category

• Planning, Routing, and Scheduling: Learning for Planning and Scheduling
• Search and Optimization: Distributed Search
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Achieving the serializable isolation level, regarded as the gold standard for transaction processing, is costly. Recent studies reveal that adjusting specific query patterns within a workload can still achieve serializability even at lower isolation levels. Nevertheless, these studies typically overlook the trade-off between the performance advantages of lower isolation levels and the overhead required to maintain serializability, potentially leading to suboptimal isolation level choices that fail to maximize performance. In this paper, we present TxnSails, a middle-tier solution designed to achieve serializable scheduling with self-adaptive isolation level selection. First, TxnSails incorporates a unified concurrency control algorithm that achieves serializability at lower isolation levels with minimal additional overhead. Second, TxnSails employs a deep learning method to characterize the trade-off between the performance benefits and overhead associated with lower isolation levels, thus predicting the optimal isolation level. Finally, TxnSails implements a cross-isolation validation mechanism to ensure serializability during real-time isolation level transitions. Extensive experiments demonstrate that TxnSails outperforms state-of-the-art solutions by up to 26.7x and PostgreSQL's serializable isolation level by up to 4.8x.

Problem

• Database Transactions
• Serializable Isolation Level
• Performance Optimization

Innovation

• Adaptive Isolation Levels
• Deep Learning Optimization
• Serializable Transactions