After Web SQL: A Cross-Engine Benchmark of IndexedDB, localStorage and SQLite over WebAssembly for Client-Side Storage

Web SQL Database, the only SQL interface ever shipped in browsers, was abandoned by the W3C in 2010 and removed from Chromium in version 119. Developers who need structured client-side storage now choose between the key–value APIs that replaced it (localStorage, IndexedDB) and SQLite compiled to WebAssembly and persisted in the Origin Private File System (OPFS). We compare six storage configurations: localStorage, IndexedDB with default and strict durability, and SQLite-Wasm 3.54.1 with an in-memory database and the two OPFS back-ends opfs and opfs-sahpool. Each configuration ran the same deterministic eight-operation workload in Chrome 157, Firefox 159 and WebKit 27.2 on 10³, 10⁴ and 10⁵ records, ten times each. All 510 runs completed and returned identical result checksums. At 10⁵ records, SQLite on opfs-sahpool loads data 5.0–7.4× faster than IndexedDB and answers a GROUP BY aggregate 8.9–13× faster. Indexed range queries and key lookups are at the same order of magnitude in both. The two OPFS back-ends differ by more than an order of magnitude: under opfs, a single-row key lookup costs 0.8–1.6 ms, 33–62× more than under opfs-sahpool. IndexedDB keeps the cheapest small write transactions (0.16–0.36 ms each) and opens 7–38× faster than SQLite-Wasm, which needs 53–60 ms just to instantiate. These results suggest that the choice of persistence layer (VFS) matters as much as the choice between SQL and key–value storage. This record contains the preprint (PDF) and the benchmark artifact (harness, raw per-run results, analysis script and figures). Text: CC BY 4.0; code: MIT. The manuscript text was drafted with AI assistance (Anthropic Claude); see the Declaration of AI Assistance in the paper.

Authors

Institutions

Publication Details

Journal
Zenodo (CERN European Organization for Nuclear Research)
Published
2026-11-22
DOI
https://doi.org/10.5281/zenodo.24430226
Primary Topic
Advanced Database Systems and Queries
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

After Web SQL: A Cross-Engine Benchmark of IndexedDB, localStorage and SQLite over WebAssembly for Client-Side Storage

Thanh Binh Le
Zenodo (CERN European Organization for Nuclear Research)
Advanced Database Systems and Queries
preprint

After Web SQL: A Cross-Engine Benchmark of IndexedDB, localStorage and SQLite over WebAssembly for Client-Side Storage

Thanh Binh Le
preprint en

Abstract

Web SQL Database, the only SQL interface ever shipped in browsers, was abandoned by the W3C in 2010 and removed from Chromium in version 119. Developers who need structured client-side storage now choose between the key–value APIs that replaced it (localStorage, IndexedDB) and SQLite compiled to WebAssembly and persisted in the Origin Private File System (OPFS). We compare six storage configurations: localStorage, IndexedDB with default and strict durability, and SQLite-Wasm 3.54.1 with an in-memory database and the two OPFS back-ends opfs and opfs-sahpool. Each configuration ran the same deterministic eight-operation workload in Chrome 157, Firefox 159 and WebKit 27.2 on 10³, 10⁴ and 10⁵ records, ten times each. All 510 runs completed and returned identical result checksums. At 10⁵ records, SQLite on opfs-sahpool loads data 5.0–7.4× faster than IndexedDB and answers a GROUP BY aggregate 8.9–13× faster. Indexed range queries and key lookups are at the same order of magnitude in both. The two OPFS back-ends differ by more than an order of magnitude: under opfs, a single-row key lookup costs 0.8–1.6 ms, 33–62× more than under opfs-sahpool. IndexedDB keeps the cheapest small write transactions (0.16–0.36 ms each) and opens 7–38× faster than SQLite-Wasm, which needs 53–60 ms just to instantiate. These results suggest that the choice of persistence layer (VFS) matters as much as the choice between SQL and key–value storage. This record contains the preprint (PDF) and the benchmark artifact (harness, raw per-run results, analysis script and figures). Text: CC BY 4.0; code: MIT. The manuscript text was drafted with AI assistance (Anthropic Claude); see the Declaration of AI Assistance in the paper.

Zenodo (CERN European Organization for Nuclear Research)
ITMO University (RU)
Advanced Database Systems and Queries
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon