Making Array-Based Translation Practical for Modern, High-Performance Buffer Management

TL;DR

This paper introduces extbf{ extcalico}, a practical buffer pool system using array-based translation for modern databases, achieving high performance across diverse workloads with innovative techniques.

Contribution

It demonstrates the viability of array-based translation in DBMS buffer pools, with new techniques to optimize performance and integrate seamlessly with existing systems.

Findings

extcalico matches or outperforms state-of-the-art in-memory and out-of-memory performance.

extcalico achieves up to 3.9× in-memory and 6.5× larger-than-memory speedup in PostgreSQL vector search.

Scan-heavy workloads see up to 3× speedup with extcalico.

Abstract

Modern buffer pools must now support a broader workload mix than classic OLTP alone. In addition to B-tree lookups, database systems increasingly serve scan-heavy analytics and vector-search indexes with irregular high-fan-out graph traversal access patterns. These workloads require a translation mechanism -- mapping logical page IDs to resident frames -- that is simultaneously fast across these diverse access patterns, deployable in user space,compatible with huge pages, easy to integrate, and still under DBMS control for eviction and I/O. Existing designs satisfy only subsets of these goals. This paper presents \textbf{\calico}, a practical DBMS-controlled buffer pool built around array-based translation, a decades-old-idea that was dissmissed but now viable with modern hardware. \calico decouples logical translation from OS page tables so that the DBMS can combine low-overhead translation with huge-page-backed frames and fine-grained page management. To make array translation practical and performant for DBMSes with large sparse hierarchical page identifiers, \calico introduces three techniques: multi-level translation with path caching, hole punching for reclaiming cold translation memory, and group prefetch to exploit parallelism. Our evaluation across scans, OLTP-style B-tree accesses, and vector search shows that \calico matches or outperforms the existing state-of-the-art in-memory and out-of-memory performance. We also implement \calico as a drop-in replacement for PostgreSQL's buffer manager and integrate it with \texttt{pgvector}. Across vector search, and scan-heavy workloads, \calico delivers up to 3.9$\times$ in-memory and 6.5$\times$ larger-than-memory speedup for PostgreSQL vector search, speeds up scan-heavy queries by up to 3$\times$.