Tiga: Accelerating Geo-Distributed Transactions with Synchronized Clocks [Technical Report]

2025-09-06

AI Summary

To address the latency-throughput trade-off inherent in strongly consistent transactions in geo-distributed databases, this paper proposes a 1-WRTT (one-wide-area-round-trip-time) commit protocol that tightly integrates concurrency control and consensus. Its core innovation is proactive transaction ordering leveraging high-precision synchronized clocks, unifying optimistic concurrency control and two-phase commit into a single WAN round trip. The protocol features an adaptive dual-path design: a fast path—based on timestamp pre-allocation and coordination-free commit—and a fallback slow path for conflict resolution. Experiments across diverse workloads demonstrate that our approach achieves 1.3×–7.2× higher throughput and 1.4×–4.6× lower end-to-end latency compared to state-of-the-art systems, while incurring significantly less computational overhead. To the best of our knowledge, this is the first protocol to realize near 1-WRTT strongly consistent transaction commits at global scale.

Technology Category

• Search and Optimization: Distributed Search
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization

Application Category

• Security and Privacy: Data transparency and provenance
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Responsible Web: Data and user privacy-enhancing technologies for the Web

Abstract

This paper presents Tiga, a new design for geo-replicated and scalable transactional databases such as Google Spanner. Tiga aims to commit transactions within 1 wide-area roundtrip time, or 1 WRTT, for a wide range of scenarios, while maintaining high throughput with minimal computational overhead. Tiga consolidates concurrency control and consensus, completing both strictly serializable execution and consistent replication in a single round. It uses synchronized clocks to proactively order transactions by assigning each a future timestamp at submission. In most cases, transactions arrive at servers before their future timestamps and are serialized according to the designated timestamp, requiring 1 WRTT to commit. In rare cases, transactions are delayed and proactive ordering fails, in which case Tiga falls back to a slow path, committing in 1.5--2 WRTTs. Compared to state-of-the-art solutions, Tiga can commit more transactions at 1-WRTT latency, and incurs much less throughput overhead. Evaluation results show that Tiga outperforms all baselines, achieving 1.3--7.2$ imes$ higher throughput and 1.4--4.6$ imes$ lower latency. Tiga is open-sourced at https://github.com/New-Consensus-Concurrency-Control/Tiga.

Problem

• Accelerating geo-distributed transactional database performance
• Reducing transaction commit latency to one roundtrip time
• Maintaining high throughput with minimal computational overhead

Innovation

• Combines concurrency control and consensus in single round
• Uses synchronized clocks for proactive transaction ordering
• Falls back to slow path for delayed transactions