GoatDB Benchmarks

A Third Category of Database

Traditional databases fall into two camps. Remote databases (PostgreSQL, Redis) store data on a server — every read and every write crosses the network. A single-row lookup costs 1–5ms in the same datacenter, 10–50ms from a browser. Embedded databases (SQLite) eliminate the network by running in-process — reads drop to ~15μs — but they offer no sync, no collaboration, and no security primitives.

GoatDB is a third option: an embedded database with built-in peer-to-peer sync whose memory model mirrors how desktop apps handle files. When your app opens a repository, the full dataset loads into memory — like a word processor opening a document. After that one-time cost, reads complete in ~1μs with no network hop and no disk I/O. The app explicitly controls which repositories are in memory by opening and closing them, the same way users open and close files. Explicit user interactions — opening a project, switching a workspace, loading a document — map directly to db.open() and db.closeRepo() calls, so the startup cost is expected and bounded. Sync, conflict resolution, and cryptographic signing run entirely in the background. The network never appears in your hot path.

The table below places GoatDB alongside the databases it draws from. The benchmark sections that follow compare GoatDB head-to-head with SQLite — the only other embedded database that runs across server, browser, and edge.

	GoatDB	SQLite	Redis	PostgreSQL
Category	Embedded, memory-first, syncing	Embedded, local-only	Remote, in-memory	Remote, disk-based
Read latency	~1μs (memory lookup)	~15μs (FFI + B-tree)	~50-200μs P50 (same datacenter)	1–5ms (network + query)
Runs in browser	Yes (OPFS)	Yes (WASM)	No	No
Built-in sync	Multi-peer CRDT	None	Pub/sub (no CRDT)	Logical replication
Offline support	Full read/write	Full read/write	None	None
Memory model	Open repos load fully into RAM, like opening a file; app explicitly controls which repos are loaded	Disk-backed, pages on demand	Entire dataset in RAM	Disk-backed, pages on demand
Data model	Commit graph with CRDT conflict resolution	B-tree + ACID transactions	Key-value / streams	Relational + ACID

Redis and PostgreSQL are not benchmarked here — they require a network hop per operation, making direct comparison misleading. Reference latency from published production data: Redis (oneuptime.com 2026, centminmod 2025), PostgreSQL (integrate.io 2026). Add 10–50ms from a browser or across regions.

Head-to-Head: GoatDB vs SQLite

Both databases in their recommended production configurations: GoatDB with cryptographic signing and relaxed durability (crash-safe via append-only log), SQLite with WAL mode and synchronous=NORMAL. Color coding: green = meaningfully faster, amber = meaningfully slower, grey = statistical tie (variance makes winner uncertain). Reproduce locally: deno task bench

Query Operations Explained

The query benchmarks test three distinct capabilities:

Filter queries are benchmarked in two configurations:

• Cold — first-access cost: GoatDB creates a new query object and performs a full O(n) scan evaluating a JavaScript predicate per item; SQLite creates a new connection and runs an equivalent WHERE clause. This represents the worst case: no caching, no reuse.
• Warm — repeated-access cost: GoatDB reuses the same live query instance (pooled by config hash) and returns its age-tracked frozen result array in O(1) when no data has changed; SQLite reuses a persistent connection with a prepared statement and a warm OS page cache. This represents steady-state performance after the first access.

The parenthetical shows dataset size and result count — e.g., "100k → 1k results" scans 100,000 items cold and returns 1,000 matches; warm, it returns the same 1,000 items directly from cache with no scan.

Filter + sort adds a custom sort comparator on top of predicate filtering (GoatDB) or ORDER BY (SQLite), measuring the combined cost. Same cold/warm split applies.

Live query updates measure reactive performance: a query is subscribed to a 100k-item dataset, then N items are modified across the predicate boundary, and the query results are refreshed. The measurement includes both write latency (N commits/UPDATEs) and query refresh time. In GoatDB, queries subscribe to changes and update incrementally via event propagation. In SQLite, the equivalent is N UPDATE statements followed by a full re-query — there is no subscription mechanism. In the browser, GoatDB queries yield cooperatively to keep the UI at 60fps, so wall-clock time includes scheduling overhead.

Server (Deno / Node.js)

Operation	GoatDB	SQLite
Create instance	1.3ms	73.7µs
Open database (empty)	437.7µs	333.6µs
Open database (100k items)	643.7ms	130.4µs
Create item	35.6µs	24.3µs
Read item	1.3µs	10.2µs
Update item	6.3µs	17.9µs
Bulk create 100 items	2.8ms	143.4µs
Bulk read 100 items	162.1µs	238.1µs
Write 100k items	2615.1ms	153.3ms
Read 100k items (cold)	278.5ms	69.9ms
Read 100k items (warm)	5.3ms ±37%	70.6ms ±91%
Filter query cold (100 items)	30.6µs	37.2µs
Filter query warm (100 items)	0.4µs	28.9µs
Filter query cold (100k → 1k results)	114.8ms	540.9µs
Filter query warm (100k → 1k results)	0.3µs	534.9µs
Filter query cold (100k → 10k results)	124.1ms	5.5ms
Filter query warm (100k → 10k results)	0.8µs	5.4ms
Filter + sort query cold (100 items)	7.0µs ±286%	17.7µs ±40%
Filter + sort query warm (100 items)	0.6µs	3.1µs
Live query update (100 items)	195.7µs	907.1µs
Live query update (1k items)	1.2ms	3.0ms
Live query update (10k items)	9.5ms	21.1ms
Count operation	1.3µs	6.5µs
Keys operation	2.9µs	11.9µs

Data: Deno. Node.js results are comparable for most operations (bulk writes diverge ~1.7×) — see Detailed Statistics for both.

Browser

GoatDB uses OPFS (Origin Private File System); SQLite uses WASM.

GoatDB writes return after committing to memory (OPFS persistence is batched in a background worker); SQLite WASM writes include synchronous OPFS persistence. This difference primarily affects single-item write latencies.

Operation	GoatDB	SQLite (WASM)
Create instance	11.9ms	10.8ms
Open database (empty)	1.8ms	2.9µs
Open database (100k items)	655.2ms	2.8ms
Create item	45.0µs	4.4ms
Read item	2.0µs	515.5µs
Update item	6.5µs	3.7ms
Bulk create 100 items	4.9ms	6.8ms
Bulk read 100 items	181.7µs	44.3ms
Read 100k items (cold)	223.1ms	378.8ms
Read 100k items (warm)	4.9ms	347.5ms
Filter query cold (100 items)	33.1µs	631.5µs
Filter query warm (100 items)	1.1µs	627.8µs
Filter query cold (100k → 1k results)	602.6ms	3.9ms
Filter query warm (100k → 1k results)	<0.1µs	3.8ms
Filter query cold (100k → 10k results)	600.7ms	35.6ms
Filter query warm (100k → 10k results)	<0.1µs	29.2ms
Filter + sort query cold (100 items)	5.6µs	498.0µs
Filter + sort query warm (100 items)	<0.1µs	469.5µs
Live query update (100 items)	170.0µs	454.2ms
Live query update (1k items)	1.4ms	4429.1ms
Live query update (10k items)	42.4ms	43846.9ms
Count operation	1.5µs	439.4µs
Keys operation	5.5µs	459.0µs

Omitted rows perform comparably across environments — see Detailed Statistics for full browser data.

Why Browser Results Differ

Three phenomena explain why the server and browser tables tell such different stories:

1. GoatDB is environment-invariant. Data lives in the JS heap regardless of runtime. Browser reads (~1.5μs) are nearly identical to server reads (~0.8μs) — both are Map lookups. The runtime boundary barely registers.

2. SQLite pays compounding browser overhead. Every browser SQLite operation crosses JS→WASM, then WASM→OPFS with synchronous I/O. On the server, SQLite uses native FFI + OS page cache. This is why read gaps jump from ~11× (server) to ~301× (browser), and live query gaps from 2.3× to 1,022×. Each individual SQLite UPDATE in the browser pays the full WASM+OPFS round-trip, so operations that issue many writes (live query updates) see the largest amplification. SQLite's :memory: mode bypasses all kernel filesystem interactions (sqlite.org/forum/forumpost/7eee1ce2e4f97f47); browser SQLite requires workarounds (Asyncify, SharedArrayBuffer+Atomics) that add 2–5× overhead (powersync.com/blog/sqlite-persistence-on-the-web).

3. Cold read direction flip. Read 100k cold is GoatDB-slower on server (273ms vs 68ms) but GoatDB-faster in browser (223ms vs 367ms). On the server, SQLite's memory-mapped pages-on-demand wins for sequential scans. In the browser, each page fetch pays the WASM+OPFS boundary cost, making pages-on-demand a liability — while GoatDB's bulk-load-and-deserialize pays the OPFS cost once.

Interpreting the Results

The database is the application cache. After the initial open, item data lives in memory — reads are pure Map lookups with no FFI, no B-tree traversal, and no disk I/O.

• Reads and metadata: GoatDB wins — ~1μs memory lookups (~11× on server, ~301× in browser). Count/keys 5–8× faster.
• Cold filter queries: SQLite wins — B-tree indexing: 2–3× faster on server (small datasets), ~18× on large. Cold = first-access cost.
• Warm filter queries: GoatDB wins — O(1) cached frozen array return. Cold-to-warm speedup: 100–1,000×. SQLite warm improvement is moderate (prepared statement + page cache).
• Live query updates: GoatDB wins — incremental event propagation vs full re-query. Browser: cooperative yielding adds scheduling overhead (not directly comparable to SQLite's blocking queries).
• Cold start: SQLite wins — pages-on-demand = near-zero open cost.

Expected outcome after running deno task bench:

• GoatDB warm reads: 10–50× faster than cold (pure Map lookups)
• GoatDB warm queries: ~100–1,000× faster than cold (O(1) cached array, no predicate scan)
• SQLite warm reads: similar to cold (B-tree traversal + FFI cost is irreducible)
• SQLite warm queries: moderate improvement over cold from prepared statement reuse and warm page cache

Cold vs. warm reads: Read 100k items (cold) measures first-access deserialization — GoatDB lazily parses each commit's JSON on the first valueForKey call, so 100k items means 100k JSON parses. Read 100k items (warm) runs one untimed full scan first to populate all caches, then times a second scan — this is GoatDB's steady-state: pure Map lookups at ~1μs each. Both databases receive identical treatment (one untimed warmup scan, then one timed scan), making the warm comparison the fairest measure of each database's in-process read throughput.

The open cost maps naturally to user-initiated actions — opening a project, switching a workspace, loading a document. Users already expect a brief load when they open something; what they don't expect is latency on every subsequent interaction. GoatDB pays the cost once at open time and then runs at memory speed, matching the mental model modern apps have trained users to expect.

Requirement	Best fit	Why
Read-heavy, long-running process	GoatDB	~1μs reads after one-time startup load; pay the cold-start cost once
Real-time reactive queries	GoatDB	Incremental query updates as data changes, no polling
Offline-first / multi-device sync	GoatDB	Built-in CRDT sync with automatic conflict resolution
Distributed AI agents	GoatDB	Runs in-process on server, edge, and browser; reactive state; optional cryptographic signing
Bulk write ingest	SQLite	B-tree + WAL optimized for batch inserts
Very large datasets (many repos open simultaneously)	SQLite / PostgreSQL	GoatDB loads each open repo fully into RAM; keep the in-memory set manageable by opening only the repos you need and closing the rest
Short-lived processes (serverless, CLI)	SQLite	Near-zero startup cost; no warm-up required
Multi-tenant server with complex queries	PostgreSQL	SQL, joins, stored procedures, row-level security
High-throughput caching layer	Redis	150K+ ops/s single instance, TTL eviction, pub/sub

Configuration Variants

The comparisons below isolate specific GoatDB configuration options. Only operations where configurations differ by more than 2× are shown.

Trusted Mode — Bypassing Signatures

GoatDB signs every commit with Ed25519 by default, enabling multi-peer and untrusted environments. For single-user or fully-trusted local scenarios you can opt out with trusted: true, which removes the signing overhead.

Operation	GoatDB	GoatDB (Trusted)
Create instance	1.3ms	846.3µs
Create item	35.6µs	15.8µs
Read item	1.3µs	0.8µs ±43%
Update item	6.3µs	2.5µs
Bulk create 100 items	2.8ms	848.3µs
Write 100k items	2615.1ms	679.2ms

Data: Deno. Only operations where trusted mode differs >1.3-1.5× — omitted rows are comparable. See Detailed Statistics for full data.

Trusted mode removes ~3× overhead on bulk writes. Reads are unaffected by signing in either mode.

Durable Mode Comparison (per-operation fsync)

GoatDB (Durable) adds per-operation fsync while keeping full cryptographic signing — isolating the cost of waiting for disk I/O. Only operations where at least one configuration differs meaningfully are shown.

Crash safety without fsync

GoatDB's append-only log remains crash-safe even without fsync — incomplete writes are discarded on restart. SQLite synchronous=OFF risks database corruption on crash. GoatDB also exposes flush() for application-controlled durability.

Operation	GoatDB (Durable)	GoatDB	SQLite	SQLite Fast-Unsafe
Create instance	821.7µs	1.3ms	73.7µs	89.2µs
Open database (empty)	316.8µs	437.7µs	333.6µs	121.8µs
Open database (100k items)	826.5ms	643.7ms	130.4µs	143.2µs
Create item	352.5µs	35.6µs	24.3µs	13.6µs
Read item	0.8µs	1.3µs	10.2µs	8.8µs
Update item	61.4µs	6.3µs	17.9µs	10.0µs
Bulk create 100 items	14.6ms	2.8ms	143.4µs	128.7µs
Bulk read 100 items	122.6µs	162.1µs	238.1µs	129.1µs
Write 100k items	2945.9ms	2615.1ms	153.3ms	140.9ms
Read 100k items (cold)	295.4ms	278.5ms	69.9ms	69.8ms
Read 100k items (warm)	6.5ms	5.3ms ±37%	70.6ms ±91%	67.3ms
Filter query cold (100 items)	10.8µs ±243%	30.6µs	37.2µs	36.4µs
Filter query warm (100 items)	0.2µs	0.4µs	28.9µs	28.7µs
Filter query cold (100k → 1k results)	113.6ms	114.8ms	540.9µs	565.6µs
Filter query warm (100k → 1k results)	0.2µs	0.3µs	534.9µs	504.9µs
Filter query cold (100k → 10k results)	122.6ms	124.1ms	5.5ms	5.6ms
Filter query warm (100k → 10k results)	0.5µs	0.8µs	5.4ms	5.5ms
Filter + sort query cold (100 items)	3.3µs	7.0µs	17.7µs	17.8µs
Filter + sort query warm (100 items)	0.3µs	0.6µs	3.1µs	2.0µs
Live query update (100 items)	175.7µs	195.7µs	907.1µs	816.4µs
Live query update (1k items)	1.3ms	1.2ms	3.0ms	3.1ms
Live query update (10k items)	11.0ms	9.5ms	21.1ms	21.0ms
Count operation	1.5µs	1.3µs	6.5µs	6.4µs
Keys operation	1.1µs	2.9µs	11.9µs	10.6µs

Data: Deno. Only operations where at least one column differs >1.3-1.5× — omitted rows are comparable. See Detailed Statistics for full data.

Storage Formats: Binary vs JSONL

GoatDB defaults to binary format (.goat, fixed-layout binary with length-prefixed framing). An optional JSONL format stores each commit as a plain-text JSON line, making the database human-readable with cat, jq, or any text editor — useful for debugging and data recovery. Only operations where the formats diverge meaningfully are shown.

Operation	Binary (default)	JSONL
Open database (empty)	437.7µs	271.8µs
Open database (100k items)	643.7ms	879.5ms
Create item	35.6µs	10.6µs
Read item	1.3µs	0.5µs
Update item	6.3µs	2.2µs
Read 100k items (cold)	278.5ms	103.2ms
Count operation	1.3µs	0.6µs
Keys operation	2.9µs	1.0µs

Data: GoatDB on Deno. Only operations where formats diverge >1.3-1.5× — omitted rows are comparable. See Detailed Statistics for full data.

The binary format (the default) produces smaller files on disk and enables fast sequential I/O. JSONL is useful for debugging — the log is human-readable with cat or jq. As the table above shows, JSONL is currently faster for some steady-state operations; binary encoding overhead is a known trade-off for compactness.

Methodology

3–10 samples per operation with warmup. Color coding: green = meaningfully faster (>1.3× ms / >1.5× μs), amber = meaningfully slower (>1.3× ms / >1.5× μs), grey = statistical tie (ranges overlap — variance makes winner uncertain), plain = within noise.

Operations with fewer than 3 samples should be treated as indicative rather than precise. In particular, live query update at 10k items runs a single iteration due to its long execution time.

Test environment

Platform	Hardware	Runtime
Deno	Apple M4 Pro, 24GB RAM	deno 2.7.5 (darwin aarch64)
Node.js	Apple M4 Pro, 24GB RAM	node v25.2.1 (darwin arm64)
Browser	Apple M4 Pro, 24GB RAM (OPFS)	browser Chrome 145.0 (Mozilla/5.0 (Macintosh; Intel Mac OS X 10_15_7) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/145.0.0.0 Safari/537.36)

Full statistics per suite

Deno

GoatDB

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	1.3ms	1.3ms	126.9µs	10%	10	781 ops/s
Open database (empty)	423.2µs	430.6µs	43.9µs	10%	7	2K ops/s
Open database (100k items)	643.7ms	648.3ms	18.4ms	3%	7	2 ops/s
Create item	35.6µs	35.9µs	10.8µs	30%	10	28K ops/s
Read item	1.3µs	1.3µs	0.3µs	25%	10	774K ops/s
Update item	6.3µs	6.3µs	1.2µs	19%	10	158K ops/s
Bulk create 100 items	2.8ms	2.7ms	116.7µs	4%	10	359 ops/s
Bulk read 100 items	166.5µs	163.4µs	18.4µs	11%	10	6K ops/s
Write 100k items	2615.1ms	2616.7ms	9.9ms	0%	7	0 ops/s
Read 100k items (cold)	278.5ms	276.9ms	5.1ms	2%	7	4 ops/s
Read 100k items (warm)	6.2ms	5.4ms	2.3ms	37%	7	162 ops/s
Filter query cold (100 items)	307.3µs	30.4µs	847.5µs	276%	10	3K ops/s
Filter query warm (100 items)	1.8µs	0.4µs	3.8µs	214%	10	561K ops/s
Filter query cold (100k → 1k results)	114.8ms	115.0ms	1.1ms	1%	10	9 ops/s
Filter query warm (100k → 1k results)	0.3µs	0.3µs	0.1µs	43%	10	2.96M ops/s
Filter query cold (100k → 10k results)	124.1ms	123.3ms	2.4ms	2%	7	8 ops/s
Filter query warm (100k → 10k results)	0.8µs	0.3µs	0.9µs	116%	7	1.28M ops/s
Filter + sort query cold (100 items)	113.3µs	6.6µs	324.0µs	286%	10	9K ops/s
Filter + sort query warm (100 items)	0.8µs	0.7µs	0.5µs	60%	10	1.22M ops/s
Live query update (100 items)	200.3µs	198.7µs	16.1µs	8%	20	5K ops/s
Live query update (1k items)	1.2ms	1.2ms	133.8µs	11%	20	842 ops/s
Live query update (10k items)	9.5ms	9.5ms	<0.1µs	0%	1	105 ops/s
Count operation	1.3µs	1.3µs	0.3µs	20%	10	777K ops/s
Keys operation	2.9µs	2.0µs	1.5µs	52%	10	351K ops/s

GoatDB (Trusted)

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	846.3µs	844.9µs	32.2µs	4%	10	1K ops/s
Open database (empty)	366.6µs	339.6µs	59.5µs	16%	7	3K ops/s
Open database (100k items)	742.1ms	740.1ms	17.7ms	2%	7	1 ops/s
Create item	15.8µs	16.2µs	3.2µs	20%	10	63K ops/s
Read item	0.8µs	0.8µs	0.4µs	43%	10	1.19M ops/s
Update item	2.5µs	2.5µs	0.8µs	33%	10	395K ops/s
Bulk create 100 items	848.3µs	827.7µs	54.5µs	6%	10	1K ops/s
Bulk read 100 items	131.8µs	131.5µs	6.7µs	5%	10	8K ops/s
Write 100k items	679.2ms	674.2ms	23.0ms	3%	7	1 ops/s
Read 100k items (cold)	291.7ms	291.1ms	1.9ms	1%	7	3 ops/s
Read 100k items (warm)	6.6ms	5.3ms	2.5ms	38%	7	152 ops/s
Filter query cold (100 items)	205.6µs	10.8µs	607.4µs	296%	10	5K ops/s
Filter query warm (100 items)	0.3µs	0.2µs	0.4µs	131%	10	3.34M ops/s
Filter query cold (100k → 1k results)	114.2ms	113.9ms	2.2ms	2%	10	9 ops/s
Filter query warm (100k → 1k results)	0.3µs	0.2µs	0.2µs	63%	10	3.58M ops/s
Filter query cold (100k → 10k results)	121.2ms	119.9ms	3.0ms	2%	7	8 ops/s
Filter query warm (100k → 10k results)	0.4µs	0.2µs	0.3µs	72%	7	2.80M ops/s
Filter + sort query cold (100 items)	36.3µs	3.4µs	95.4µs	263%	10	28K ops/s
Filter + sort query warm (100 items)	0.4µs	0.2µs	0.3µs	86%	10	2.73M ops/s
Live query update (100 items)	168.3µs	168.5µs	7.3µs	4%	20	6K ops/s
Live query update (1k items)	972.1µs	964.5µs	125.7µs	13%	20	1K ops/s
Live query update (10k items)	10.4ms	10.4ms	<0.1µs	0%	1	96 ops/s
Count operation	1.0µs	1.0µs	0.2µs	20%	10	984K ops/s
Keys operation	1.9µs	1.5µs	0.8µs	43%	10	524K ops/s

GoatDB (Durable)

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	849.3µs	814.7µs	95.1µs	11%	10	1K ops/s
Open database (empty)	316.8µs	329.8µs	24.7µs	8%	7	3K ops/s
Open database (100k items)	826.5ms	827.1ms	19.4ms	2%	7	1 ops/s
Create item	359.0µs	359.7µs	41.8µs	12%	10	3K ops/s
Read item	1.1µs	0.9µs	0.6µs	54%	10	923K ops/s
Update item	63.5µs	59.8µs	9.8µs	15%	10	16K ops/s
Bulk create 100 items	14.6ms	15.6ms	2.1ms	14%	10	68 ops/s
Bulk read 100 items	122.6µs	121.4µs	2.4µs	2%	10	8K ops/s
Write 100k items	2945.9ms	2946.4ms	12.1ms	0%	7	0 ops/s
Read 100k items (cold)	295.4ms	296.6ms	3.8ms	1%	7	3 ops/s
Read 100k items (warm)	6.5ms	5.5ms	2.2ms	34%	7	155 ops/s
Filter query cold (100 items)	58.5µs	10.5µs	141.9µs	243%	10	17K ops/s
Filter query warm (100 items)	0.3µs	0.2µs	0.4µs	128%	10	3.16M ops/s
Filter query cold (100k → 1k results)	113.6ms	113.9ms	1.2ms	1%	10	9 ops/s
Filter query warm (100k → 1k results)	0.2µs	0.2µs	0.1µs	37%	10	4.53M ops/s
Filter query cold (100k → 10k results)	122.6ms	121.8ms	3.1ms	3%	7	8 ops/s
Filter query warm (100k → 10k results)	0.5µs	0.3µs	0.5µs	94%	7	1.95M ops/s
Filter + sort query cold (100 items)	53.6µs	3.0µs	150.2µs	280%	10	19K ops/s
Filter + sort query warm (100 items)	0.4µs	0.2µs	0.4µs	104%	10	2.67M ops/s
Live query update (100 items)	177.1µs	177.0µs	9.5µs	5%	20	6K ops/s
Live query update (1k items)	1.3ms	1.3ms	81.1µs	6%	20	795 ops/s
Live query update (10k items)	11.0ms	11.0ms	<0.1µs	0%	1	91 ops/s
Count operation	3.2µs	1.4µs	4.0µs	126%	10	315K ops/s
Keys operation	1.1µs	1.1µs	0.1µs	11%	10	873K ops/s

GoatDB JSONL

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	918.4µs	907.8µs	114.7µs	12%	10	1K ops/s
Open database (empty)	271.8µs	274.5µs	22.3µs	8%	7	4K ops/s
Open database (100k items)	929.9ms	875.0ms	138.1ms	15%	7	1 ops/s
Create item	11.8µs	10.9µs	3.9µs	33%	10	85K ops/s
Read item	0.5µs	0.5µs	0.1µs	26%	10	2.00M ops/s
Update item	2.2µs	2.1µs	0.5µs	25%	10	455K ops/s
Bulk create 100 items	2.6ms	2.6ms	48.2µs	2%	10	383 ops/s
Bulk read 100 items	130.4µs	125.8µs	10.4µs	8%	10	8K ops/s
Write 100k items	2668.8ms	2667.1ms	10.7ms	0%	7	0 ops/s
Read 100k items (cold)	103.2ms	101.8ms	4.3ms	4%	7	10 ops/s
Read 100k items (warm)	6.9ms	5.2ms	3.0ms	43%	7	145 ops/s
Filter query cold (100 items)	285.7µs	10.5µs	855.5µs	299%	10	3K ops/s
Filter query warm (100 items)	0.2µs	0.1µs	0.2µs	110%	10	6.00M ops/s
Filter query cold (100k → 1k results)	115.9ms	116.0ms	1.4ms	1%	10	9 ops/s
Filter query warm (100k → 1k results)	0.2µs	0.1µs	0.1µs	95%	10	6.66M ops/s
Filter query cold (100k → 10k results)	120.0ms	120.5ms	2.3ms	2%	7	8 ops/s
Filter query warm (100k → 10k results)	0.4µs	0.3µs	0.5µs	117%	7	2.40M ops/s
Filter + sort query cold (100 items)	162.8µs	4.3µs	489.7µs	301%	10	6K ops/s
Filter + sort query warm (100 items)	0.2µs	0.1µs	0.1µs	67%	10	5.70M ops/s
Live query update (100 items)	182.0µs	180.8µs	12.4µs	7%	20	5K ops/s
Live query update (1k items)	1.2ms	1.2ms	47.0µs	4%	20	848 ops/s
Live query update (10k items)	11.1ms	11.1ms	<0.1µs	0%	1	90 ops/s
Count operation	0.6µs	0.6µs	0.1µs	21%	10	1.63M ops/s
Keys operation	1.0µs	0.9µs	0.2µs	21%	10	1.04M ops/s

SQLite

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	77.1µs	74.9µs	7.6µs	10%	10	13K ops/s
Open database (empty)	333.6µs	314.3µs	34.7µs	10%	7	3K ops/s
Open database (100k items)	130.4µs	124.8µs	20.7µs	16%	7	8K ops/s
Create item	26.5µs	24.6µs	5.0µs	19%	10	38K ops/s
Read item	10.2µs	9.5µs	2.0µs	20%	10	98K ops/s
Update item	17.9µs	18.0µs	1.0µs	6%	10	56K ops/s
Bulk create 100 items	143.4µs	143.0µs	3.1µs	2%	10	7K ops/s
Bulk read 100 items	238.1µs	231.9µs	14.9µs	6%	10	4K ops/s
Write 100k items	153.3ms	150.8ms	3.8ms	3%	7	7 ops/s
Read 100k items (cold)	69.9ms	68.0ms	5.0ms	7%	7	14 ops/s
Read 100k items (warm)	107.3ms	69.0ms	97.5ms	91%	7	9 ops/s
Filter query cold (100 items)	38.7µs	36.1µs	5.4µs	14%	10	26K ops/s
Filter query warm (100 items)	29.2µs	28.8µs	1.0µs	4%	10	34K ops/s
Filter query cold (100k → 1k results)	543.3µs	540.1µs	9.4µs	2%	10	2K ops/s
Filter query warm (100k → 1k results)	539.5µs	534.6µs	17.2µs	3%	10	2K ops/s
Filter query cold (100k → 10k results)	5.6ms	5.5ms	486.3µs	9%	10	178 ops/s
Filter query warm (100k → 10k results)	5.5ms	5.4ms	444.3µs	8%	10	181 ops/s
Filter + sort query cold (100 items)	20.0µs	17.0µs	8.0µs	40%	10	50K ops/s
Filter + sort query warm (100 items)	3.3µs	3.0µs	0.8µs	23%	10	300K ops/s
Live query update (100 items)	982.8µs	919.3µs	305.5µs	31%	20	1K ops/s
Live query update (1k items)	3.0ms	3.2ms	240.9µs	8%	20	329 ops/s
Live query update (10k items)	21.1ms	20.9ms	2.8ms	13%	20	47 ops/s
Count operation	7.1µs	6.3µs	2.2µs	31%	10	140K ops/s
Keys operation	11.9µs	11.3µs	1.4µs	12%	10	84K ops/s

SQLite Fast-Unsafe

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	89.2µs	84.4µs	13.8µs	16%	10	11K ops/s
Open database (empty)	121.8µs	122.7µs	9.6µs	8%	7	8K ops/s
Open database (100k items)	143.2µs	142.0µs	8.5µs	6%	7	7K ops/s
Create item	15.3µs	13.7µs	6.4µs	42%	10	65K ops/s
Read item	8.8µs	8.0µs	2.4µs	27%	10	114K ops/s
Update item	10.0µs	9.9µs	1.3µs	13%	10	100K ops/s
Bulk create 100 items	128.7µs	129.6µs	2.6µs	2%	10	8K ops/s
Bulk read 100 items	129.1µs	128.1µs	6.4µs	5%	10	8K ops/s
Write 100k items	140.9ms	140.5ms	2.1ms	2%	7	7 ops/s
Read 100k items (cold)	69.8ms	69.4ms	5.4ms	8%	7	14 ops/s
Read 100k items (warm)	67.3ms	66.8ms	4.1ms	6%	7	15 ops/s
Filter query cold (100 items)	36.4µs	35.9µs	1.9µs	5%	10	27K ops/s
Filter query warm (100 items)	28.7µs	27.5µs	1.9µs	6%	10	35K ops/s
Filter query cold (100k → 1k results)	572.8µs	565.5µs	25.2µs	4%	10	2K ops/s
Filter query warm (100k → 1k results)	501.7µs	503.5µs	13.1µs	3%	10	2K ops/s
Filter query cold (100k → 10k results)	6.0ms	5.7ms	763.2µs	13%	10	167 ops/s
Filter query warm (100k → 10k results)	5.8ms	5.5ms	731.8µs	13%	10	172 ops/s
Filter + sort query cold (100 items)	19.1µs	17.9µs	5.2µs	27%	10	52K ops/s
Filter + sort query warm (100 items)	2.2µs	1.9µs	0.8µs	35%	10	454K ops/s
Live query update (100 items)	850.6µs	832.7µs	81.4µs	10%	20	1K ops/s
Live query update (1k items)	3.1ms	3.1ms	325.7µs	11%	20	326 ops/s
Live query update (10k items)	21.0ms	20.7ms	3.2ms	15%	20	48 ops/s
Count operation	6.4µs	6.2µs	1.3µs	20%	10	155K ops/s
Keys operation	10.6µs	10.1µs	1.6µs	15%	10	94K ops/s

Node.js

GoatDB

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	934.5µs	915.3µs	106.7µs	11%	10	1K ops/s
Open database (empty)	293.6µs	287.3µs	24.8µs	8%	7	3K ops/s
Open database (100k items)	666.5ms	663.1ms	16.2ms	2%	7	2 ops/s
Create item	29.7µs	28.8µs	9.0µs	30%	10	34K ops/s
Read item	0.9µs	0.9µs	0.3µs	30%	10	1.12M ops/s
Update item	5.5µs	4.9µs	2.0µs	36%	10	181K ops/s
Bulk create 100 items	4.4ms	4.3ms	291.9µs	7%	10	226 ops/s
Bulk read 100 items	158.5µs	155.0µs	16.0µs	10%	10	6K ops/s
Write 100k items	4504.2ms	4502.5ms	33.0ms	1%	7	0 ops/s
Read 100k items (cold)	300.9ms	301.6ms	3.9ms	1%	7	3 ops/s
Read 100k items (warm)	8.5ms	7.8ms	1.8ms	21%	7	118 ops/s
Filter query cold (100 items)	174.4µs	30.8µs	433.7µs	249%	10	6K ops/s
Filter query warm (100 items)	1.7µs	0.6µs	3.7µs	216%	10	590K ops/s
Filter query cold (100k → 1k results)	117.6ms	117.9ms	4.0ms	3%	10	9 ops/s
Filter query warm (100k → 1k results)	0.4µs	0.2µs	0.2µs	58%	10	2.79M ops/s
Filter query cold (100k → 10k results)	126.5ms	127.3ms	3.2ms	3%	7	8 ops/s
Filter query warm (100k → 10k results)	1.4µs	0.6µs	2.5µs	174%	7	706K ops/s
Filter + sort query cold (100 items)	55.7µs	7.3µs	144.3µs	259%	10	18K ops/s
Filter + sort query warm (100 items)	0.4µs	0.3µs	0.3µs	64%	10	2.22M ops/s
Live query update (100 items)	194.4µs	189.9µs	19.5µs	10%	20	5K ops/s
Live query update (1k items)	1.1ms	1.1ms	67.9µs	6%	20	910 ops/s
Live query update (10k items)	7.3ms	7.3ms	<0.1µs	0%	1	138 ops/s
Count operation	1.3µs	1.2µs	0.4µs	34%	10	760K ops/s
Keys operation	2.5µs	2.1µs	1.0µs	41%	10	403K ops/s

GoatDB (Trusted)

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	805.9µs	791.2µs	52.5µs	7%	10	1K ops/s
Open database (empty)	351.9µs	298.5µs	126.0µs	36%	7	3K ops/s
Open database (100k items)	676.4ms	680.2ms	29.2ms	4%	7	1 ops/s
Create item	16.7µs	11.9µs	12.3µs	74%	10	60K ops/s
Read item	0.6µs	0.5µs	0.2µs	43%	10	1.73M ops/s
Update item	1.9µs	1.9µs	0.5µs	27%	10	513K ops/s
Bulk create 100 items	732.8µs	715.7µs	56.5µs	8%	10	1K ops/s
Bulk read 100 items	130.0µs	125.9µs	11.5µs	9%	10	8K ops/s
Write 100k items	793.5ms	742.1ms	99.5ms	13%	7	1 ops/s
Read 100k items (cold)	317.0ms	313.1ms	8.9ms	3%	7	3 ops/s
Read 100k items (warm)	8.4ms	7.7ms	1.9ms	23%	7	119 ops/s
Filter query cold (100 items)	63.7µs	14.6µs	147.7µs	232%	10	16K ops/s
Filter query warm (100 items)	0.5µs	0.4µs	0.4µs	83%	10	2.07M ops/s
Filter query cold (100k → 1k results)	120.5ms	120.9ms	2.8ms	2%	10	8 ops/s
Filter query warm (100k → 1k results)	0.3µs	0.3µs	0.2µs	45%	10	3.00M ops/s
Filter query cold (100k → 10k results)	129.6ms	128.4ms	4.9ms	4%	7	8 ops/s
Filter query warm (100k → 10k results)	0.8µs	0.4µs	0.9µs	106%	7	1.24M ops/s
Filter + sort query cold (100 items)	37.3µs	4.0µs	96.1µs	257%	10	27K ops/s
Filter + sort query warm (100 items)	0.4µs	0.2µs	0.4µs	97%	10	2.45M ops/s
Live query update (100 items)	175.3µs	174.5µs	15.3µs	9%	20	6K ops/s
Live query update (1k items)	878.7µs	870.7µs	78.3µs	9%	20	1K ops/s
Live query update (10k items)	7.5ms	7.5ms	<0.1µs	0%	1	134 ops/s
Count operation	1.2µs	1.2µs	0.2µs	16%	10	848K ops/s
Keys operation	3.1µs	1.4µs	5.0µs	163%	10	326K ops/s

GoatDB (Durable)

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	831.9µs	804.9µs	95.2µs	11%	10	1K ops/s
Open database (empty)	266.8µs	264.0µs	14.5µs	5%	7	4K ops/s
Open database (100k items)	708.1ms	707.4ms	18.2ms	3%	7	1 ops/s
Create item	310.0µs	310.9µs	20.5µs	7%	10	3K ops/s
Read item	0.6µs	0.5µs	0.2µs	39%	10	1.77M ops/s
Update item	73.9µs	71.4µs	8.7µs	12%	10	14K ops/s
Bulk create 100 items	12.9ms	12.7ms	681.1µs	5%	10	78 ops/s
Bulk read 100 items	123.9µs	123.2µs	4.3µs	3%	10	8K ops/s
Write 100k items	4703.1ms	4710.1ms	99.6ms	2%	7	0 ops/s
Read 100k items (cold)	345.8ms	349.2ms	26.5ms	8%	7	3 ops/s
Read 100k items (warm)	9.9ms	9.2ms	2.2ms	22%	7	101 ops/s
Filter query cold (100 items)	72.3µs	16.9µs	136.9µs	189%	10	14K ops/s
Filter query warm (100 items)	0.4µs	0.2µs	0.5µs	125%	10	2.53M ops/s
Filter query cold (100k → 1k results)	128.8ms	129.0ms	3.8ms	3%	10	8 ops/s
Filter query warm (100k → 1k results)	0.4µs	0.4µs	0.2µs	53%	10	2.47M ops/s
Filter query cold (100k → 10k results)	136.5ms	136.6ms	4.6ms	3%	7	7 ops/s
Filter query warm (100k → 10k results)	0.4µs	0.3µs	0.3µs	67%	7	2.44M ops/s
Filter + sort query cold (100 items)	36.6µs	4.1µs	97.2µs	265%	10	27K ops/s
Filter + sort query warm (100 items)	0.4µs	0.2µs	0.4µs	112%	10	2.64M ops/s
Live query update (100 items)	187.0µs	186.5µs	11.7µs	6%	20	5K ops/s
Live query update (1k items)	1.4ms	1.1ms	1.4ms	103%	20	714 ops/s
Live query update (10k items)	7.5ms	7.5ms	<0.1µs	0%	1	133 ops/s
Count operation	2.7µs	1.9µs	2.5µs	94%	10	374K ops/s
Keys operation	1.1µs	1.1µs	0.2µs	16%	10	909K ops/s

GoatDB JSONL

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	848.6µs	794.7µs	152.6µs	18%	10	1K ops/s
Open database (empty)	256.1µs	261.5µs	15.2µs	6%	7	4K ops/s
Open database (100k items)	873.8ms	859.5ms	76.1ms	9%	7	1 ops/s
Create item	11.0µs	10.6µs	1.8µs	16%	10	91K ops/s
Read item	0.4µs	0.3µs	0.3µs	66%	10	2.47M ops/s
Update item	2.5µs	2.7µs	0.4µs	17%	10	405K ops/s
Bulk create 100 items	4.5ms	4.3ms	613.2µs	14%	10	223 ops/s
Bulk read 100 items	120.2µs	118.9µs	3.3µs	3%	10	8K ops/s
Write 100k items	4804.1ms	4779.3ms	108.0ms	2%	7	0 ops/s
Read 100k items (cold)	155.8ms	148.2ms	16.7ms	11%	7	6 ops/s
Read 100k items (warm)	9.3ms	8.3ms	2.0ms	22%	7	107 ops/s
Filter query cold (100 items)	60.0µs	14.5µs	137.5µs	229%	10	17K ops/s
Filter query warm (100 items)	0.2µs	0.1µs	0.2µs	101%	10	6.15M ops/s
Filter query cold (100k → 1k results)	129.6ms	130.0ms	3.4ms	3%	10	8 ops/s
Filter query warm (100k → 1k results)	0.1µs	0.1µs	0.1µs	66%	10	7.75M ops/s
Filter query cold (100k → 10k results)	133.1ms	134.4ms	4.9ms	4%	7	8 ops/s
Filter query warm (100k → 10k results)	0.4µs	0.4µs	0.4µs	88%	7	2.33M ops/s
Filter + sort query cold (100 items)	38.1µs	4.6µs	99.9µs	262%	10	26K ops/s
Filter + sort query warm (100 items)	0.2µs	0.1µs	0.2µs	108%	10	5.11M ops/s
Live query update (100 items)	172.4µs	167.4µs	21.2µs	12%	20	6K ops/s
Live query update (1k items)	1.3ms	1.3ms	68.2µs	5%	20	754 ops/s
Live query update (10k items)	7.5ms	7.5ms	<0.1µs	0%	1	134 ops/s
Count operation	0.8µs	0.7µs	0.2µs	27%	10	1.20M ops/s
Keys operation	1.6µs	1.1µs	1.6µs	103%	10	638K ops/s

SQLite

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	82.6µs	81.1µs	13.8µs	17%	10	12K ops/s
Open database (empty)	332.4µs	316.1µs	40.7µs	12%	7	3K ops/s
Open database (100k items)	98.5µs	99.7µs	6.2µs	6%	7	10K ops/s
Create item	23.2µs	22.9µs	2.1µs	9%	10	43K ops/s
Read item	8.0µs	6.8µs	2.1µs	26%	10	125K ops/s
Update item	18.4µs	17.6µs	3.5µs	19%	10	54K ops/s
Bulk create 100 items	110.1µs	108.0µs	3.8µs	3%	10	9K ops/s
Bulk read 100 items	194.7µs	193.0µs	7.7µs	4%	10	5K ops/s
Write 100k items	117.1ms	114.7ms	4.0ms	3%	7	9 ops/s
Read 100k items (cold)	38.4ms	39.5ms	2.7ms	7%	7	26 ops/s
Read 100k items (warm)	37.2ms	36.6ms	1.7ms	4%	7	27 ops/s
Filter query cold (100 items)	23.7µs	23.6µs	0.8µs	3%	10	42K ops/s
Filter query warm (100 items)	17.3µs	17.0µs	0.8µs	5%	10	58K ops/s
Filter query cold (100k → 1k results)	331.1µs	330.4µs	8.8µs	3%	10	3K ops/s
Filter query warm (100k → 1k results)	310.1µs	306.1µs	20.5µs	7%	10	3K ops/s
Filter query cold (100k → 10k results)	3.4ms	3.2ms	760.4µs	22%	10	295 ops/s
Filter query warm (100k → 10k results)	3.2ms	3.0ms	727.2µs	23%	10	312 ops/s
Filter + sort query cold (100 items)	13.5µs	12.0µs	3.2µs	24%	10	74K ops/s
Filter + sort query warm (100 items)	2.7µs	2.4µs	0.8µs	28%	10	366K ops/s
Live query update (100 items)	601.9µs	595.3µs	34.7µs	6%	20	2K ops/s
Live query update (1k items)	2.2ms	2.2ms	135.6µs	6%	20	453 ops/s
Live query update (10k items)	15.0ms	14.9ms	1.5ms	10%	20	67 ops/s
Count operation	6.4µs	6.5µs	0.9µs	14%	10	157K ops/s
Keys operation	9.4µs	8.8µs	1.4µs	15%	10	107K ops/s

SQLite Fast-Unsafe

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	74.3µs	73.7µs	5.8µs	8%	10	13K ops/s
Open database (empty)	102.3µs	102.0µs	4.4µs	4%	7	10K ops/s
Open database (100k items)	125.3µs	125.5µs	6.5µs	5%	7	8K ops/s
Create item	9.8µs	9.7µs	1.5µs	16%	10	102K ops/s
Read item	7.2µs	6.5µs	2.2µs	31%	10	138K ops/s
Update item	7.3µs	7.2µs	1.0µs	13%	10	137K ops/s
Bulk create 100 items	91.5µs	91.0µs	2.1µs	2%	10	11K ops/s
Bulk read 100 items	112.8µs	111.6µs	4.5µs	4%	10	9K ops/s
Write 100k items	104.9ms	103.5ms	2.4ms	2%	7	10 ops/s
Read 100k items (cold)	37.2ms	36.5ms	3.7ms	10%	7	27 ops/s
Read 100k items (warm)	36.2ms	36.2ms	501.5µs	1%	7	28 ops/s
Filter query cold (100 items)	22.6µs	22.1µs	1.3µs	6%	10	44K ops/s
Filter query warm (100 items)	15.1µs	14.5µs	1.6µs	10%	10	66K ops/s
Filter query cold (100k → 1k results)	311.8µs	312.7µs	5.9µs	2%	10	3K ops/s
Filter query warm (100k → 1k results)	313.7µs	304.8µs	27.6µs	9%	10	3K ops/s
Filter query cold (100k → 10k results)	3.1ms	3.1ms	45.0µs	1%	10	323 ops/s
Filter query warm (100k → 10k results)	3.0ms	3.0ms	17.8µs	1%	10	336 ops/s
Filter + sort query cold (100 items)	13.8µs	12.5µs	3.5µs	26%	10	73K ops/s
Filter + sort query warm (100 items)	1.8µs	1.5µs	0.8µs	43%	10	543K ops/s
Live query update (100 items)	518.8µs	519.8µs	21.4µs	4%	20	2K ops/s
Live query update (1k items)	2.2ms	2.1ms	425.0µs	20%	20	463 ops/s
Live query update (10k items)	14.5ms	14.5ms	1.5ms	10%	20	69 ops/s
Count operation	5.0µs	4.6µs	1.3µs	25%	10	200K ops/s
Keys operation	9.0µs	7.6µs	3.8µs	43%	10	112K ops/s

Browser

GoatDB

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	11.9ms	11.8ms	652.7µs	5%	10	84 ops/s
Open database (empty)	1.8ms	1.7ms	357.4µs	20%	7	551 ops/s
Open database (100k items)	655.2ms	662.2ms	20.2ms	3%	7	2 ops/s
Create item	45.0µs	45.0µs	12.7µs	28%	10	22K ops/s
Read item	2.0µs	<0.1µs	2.6µs	129%	10	500K ops/s
Update item	6.5µs	5.0µs	3.4µs	52%	10	154K ops/s
Bulk create 100 items	4.9ms	4.9ms	277.1µs	6%	10	204 ops/s
Bulk read 100 items	185.0µs	180.0µs	12.9µs	7%	10	5K ops/s
Read 100k items (cold)	223.1ms	223.5ms	3.3ms	1%	7	4 ops/s
Read 100k items (warm)	5.4ms	4.7ms	1.2ms	22%	7	186 ops/s
Filter query cold (100 items)	260.0µs	32.5µs	699.5µs	269%	10	4K ops/s
Filter query warm (100 items)	2.5µs	<0.1µs	4.9µs	194%	10	400K ops/s
Filter query cold (100k → 1k results)	600.7ms	600.7ms	8.4ms	1%	10	2 ops/s
Filter query warm (100k → 1k results)	<0.1µs	<0.1µs	<0.1µs	0%	10	—
Filter query cold (100k → 10k results)	600.7ms	600.9ms	6.6ms	1%	7	2 ops/s
Filter query warm (100k → 10k results)	<0.1µs	<0.1µs	<0.1µs	0%	7	—
Filter + sort query cold (100 items)	172.5µs	5.0µs	520.9µs	302%	10	6K ops/s
Filter + sort query warm (100 items)	1.0µs	<0.1µs	2.1µs	211%	10	1.00M ops/s
Live query update (100 items)	173.3µs	165.0µs	25.0µs	14%	20	6K ops/s
Live query update (1k items)	1.4ms	1.4ms	81.1µs	6%	20	710 ops/s
Live query update (10k items)	42.4ms	42.4ms	<0.1µs	0%	1	24 ops/s
Count operation	1.5µs	<0.1µs	2.4µs	161%	10	667K ops/s
Keys operation	5.5µs	5.0µs	4.4µs	80%	10	182K ops/s

GoatDB JSONL

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	12.6ms	12.2ms	1.4ms	11%	10	79 ops/s
Open database (empty)	2.3ms	2.2ms	917.7µs	39%	7	430 ops/s
Open database (100k items)	1236.0ms	1258.7ms	89.1ms	7%	7	1 ops/s
Create item	23.5µs	25.0µs	4.1µs	18%	10	43K ops/s
Read item	0.5µs	<0.1µs	1.6µs	316%	10	2.00M ops/s
Update item	5.0µs	5.0µs	2.4µs	47%	10	200K ops/s
Bulk create 100 items	4.8ms	4.7ms	243.9µs	5%	10	210 ops/s
Bulk read 100 items	154.0µs	155.0µs	8.1µs	5%	10	6K ops/s
Read 100k items (cold)	94.2ms	83.7ms	27.6ms	29%	7	11 ops/s
Read 100k items (warm)	5.2ms	4.8ms	974.8µs	19%	7	194 ops/s
Filter query cold (100 items)	191.5µs	15.0µs	552.9µs	289%	10	5K ops/s
Filter query warm (100 items)	0.5µs	<0.1µs	1.6µs	316%	10	2.00M ops/s
Filter query cold (100k → 1k results)	601.2ms	600.0ms	4.5ms	1%	10	2 ops/s
Filter query warm (100k → 1k results)	0.5µs	<0.1µs	1.6µs	316%	10	2.00M ops/s
Filter query cold (100k → 10k results)	600.4ms	599.4ms	5.2ms	1%	7	2 ops/s
Filter query warm (100k → 10k results)	<0.1µs	<0.1µs	<0.1µs	0%	7	—
Filter + sort query cold (100 items)	186.5µs	5.0µs	568.8µs	305%	10	5K ops/s
Filter + sort query warm (100 items)	1.5µs	<0.1µs	2.4µs	161%	10	667K ops/s
Live query update (100 items)	144.5µs	145.0µs	7.2µs	5%	20	7K ops/s
Live query update (1k items)	1.2ms	1.2ms	53.7µs	4%	20	825 ops/s
Live query update (10k items)	43.8ms	43.8ms	<0.1µs	0%	1	23 ops/s
Count operation	2.0µs	<0.1µs	3.5µs	175%	10	500K ops/s
Keys operation	2.0µs	<0.1µs	2.6µs	129%	10	500K ops/s

SQLite

Operation	Average	Median	Stddev	CV	Samples	Throughput
Create instance	10.8ms	10.8ms	904.5µs	8%	10	93 ops/s
Open database (empty)	2.9µs	5.0µs	2.7µs	94%	7	350K ops/s
Open database (100k items)	2.8ms	2.8ms	96.2µs	3%	7	357 ops/s
Create item	4.4ms	4.4ms	172.7µs	4%	10	229 ops/s
Read item	515.5µs	497.5µs	61.5µs	12%	10	2K ops/s
Update item	3.7ms	3.7ms	178.3µs	5%	10	269 ops/s
Bulk create 100 items	6.8ms	6.8ms	161.3µs	2%	10	146 ops/s
Bulk read 100 items	45.7ms	44.7ms	3.3ms	7%	10	22 ops/s
Read 100k items (cold)	378.8ms	380.0ms	3.9ms	1%	7	3 ops/s
Read 100k items (warm)	347.5ms	346.3ms	10.7ms	3%	7	3 ops/s
Filter query cold (100 items)	631.5µs	635.0µs	31.4µs	5%	10	2K ops/s
Filter query warm (100 items)	812.0µs	625.0µs	584.9µs	72%	10	1K ops/s
Filter query cold (100k → 1k results)	3.9ms	3.9ms	106.7µs	3%	10	254 ops/s
Filter query warm (100k → 1k results)	3.8ms	3.1ms	1.3ms	35%	10	261 ops/s
Filter query cold (100k → 10k results)	35.6ms	35.3ms	828.4µs	2%	10	28 ops/s
Filter query warm (100k → 10k results)	29.2ms	28.3ms	1.8ms	6%	10	34 ops/s
Filter + sort query cold (100 items)	498.0µs	487.5µs	93.8µs	19%	10	2K ops/s
Filter + sort query warm (100 items)	469.5µs	465.0µs	35.1µs	7%	10	2K ops/s
Live query update (100 items)	474.5ms	455.9ms	68.7ms	14%	20	2 ops/s
Live query update (1k items)	4432.8ms	4429.7ms	42.7ms	1%	20	0 ops/s
Live query update (10k items)	43846.9ms	43796.0ms	205.1ms	0%	20	—
Count operation	446.0µs	442.5µs	24.2µs	5%	10	2K ops/s
Keys operation	459.0µs	455.0µs	17.9µs	4%	10	2K ops/s

Running Your Own Benchmarks

git clone https://github.com/goatplatform/goatdb
cd goatdb
deno task bench

The benchmark suite runs on Deno, Node.js, and browser automatically.