TierBase: A Workload-Driven Cost-Optimized Key-Value Store

2025-05-10

AI Summary

Data-intensive online services face a critical trade-off between storage cost and performance, particularly when managing heterogeneous memory tiers (cache, persistent memory [PMem], and disk). Method: This paper proposes a workload-driven joint space-performance cost model to orchestrate multi-tier storage hierarchies. It innovatively integrates pretrained lightweight compression, an elastic thread pool, and a PMem-aware hierarchical synchronization protocol—enhancing resource utilization and robustness against workload skew. Contribution/Results: Evaluated in real production environments, the system reduces total storage ownership cost by up to 62% versus baseline approaches. At equivalent performance levels, its end-to-end cost is 37–58% lower than mainstream key-value systems, significantly outperforming state-of-the-art solutions in both cost efficiency and adaptability.

Technology Category

• Data Mining & Knowledge Management: Scalability, Parallel & Distributed Systems
• Machine Learning: Hardware-aware ML
• Search and Optimization: Distributed Search

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

In the current era of data-intensive applications, the demand for high-performance, cost-effective storage solutions is paramount. This paper introduces a Space-Performance Cost Model for key-value store, designed to guide cost-effective storage configuration decisions. The model quantifies the trade-offs between performance and storage costs, providing a framework for optimizing resource allocation in large-scale data serving environments. Guided by this cost model, we present TierBase, a distributed key-value store developed by Ant Group that optimizes total cost by strategically synchronizing data between cache and storage tiers, maximizing resource utilization and effectively handling skewed workloads. To enhance cost-efficiency, TierBase incorporates several optimization techniques, including pre-trained data compression, elastic threading mechanisms, and the utilization of persistent memory. We detail TierBase's architecture, key components, and the implementation of cost optimization strategies. Extensive evaluations using both synthetic benchmarks and real-world workloads demonstrate TierBase's superior cost-effectiveness compared to existing solutions. Furthermore, case studies from Ant Group's production environments showcase TierBase's ability to achieve up to 62% cost reduction in primary scenarios, highlighting its practical impact in large-scale online data serving.

Problem

• Optimizing cost-performance trade-offs in key-value stores
• Designing workload-driven storage configurations for large-scale data
• Reducing operational costs while handling skewed workloads effectively

Innovation

• Space-Performance Cost Model for key-value stores
• Strategic data synchronization between cache and storage
• Pre-trained data compression and elastic threading