KalamDB Benchmark Report

Performance analysis — v0.4.1-beta

http://localhost:8080 2026-03-13T13:18:45.938287+00:00 100 iters 5 warmup 10 concurrency

Total Benchmarks

Passed

Failed

785.95s

Total Duration

System Information

Hostname

Jamals-MacBook-Pro.local

Machine Model

MacBookPro18,3

CPU Model

Apple M1 Pro

CPU Cores

10 logical / 10 physical

Total Memory

16.00 GiB

Available Memory

4.44 GiB

Used Memory

11.56 GiB

Memory Usage

72.2%

Darwin macOS 26.4

Kernel

25.4.0

Architecture

aarch64

Latency by Operation (µs)

Throughput (ops/sec)

Avg Latency by Category (µs)

Detailed Results

Status	Benchmark	Category	Description	Iters	Mean	P50	P95	P99	Min	Max	Ops/sec	Total	Verdict	vs Prev
PASS	create_table	DDL	CREATE TABLE with 3 columns	100	38.4ms	39.7ms	59.9ms	61.5ms	12.8ms	61.7ms	26	3.84s	🟢 Excellent	↑25% faster
PASS	drop_table	DDL	DROP TABLE on a previously created table	100	107.7ms	107.8ms	175.3ms	184.4ms	30.7ms	184.7ms	9	10.77s	🟡 Acceptable	↓116% slower
PASS	single_insert	Insert	INSERT a single row into a table	100	466µs	374µs	725µs	2.6ms	292µs	3.8ms	2.1K	46.6ms	🟢 Excellent	↓66% slower
PASS	bulk_insert	Insert	INSERT 50 rows in a single statement	100	1.1ms	1.0ms	1.4ms	2.3ms	868µs	2.4ms	913	109.5ms	🟢 Excellent	↓69% slower
PASS	select_all	Select	SELECT * from a 200-row table	100	1.6ms	1.5ms	2.0ms	3.4ms	1.3ms	3.7ms	615	162.7ms	🟢 Excellent	↓42% slower
PASS	select_by_filter	Select	SELECT with WHERE clause on a 200-row table	100	709µs	656µs	924µs	1.8ms	492µs	2.6ms	1.4K	70.9ms	🟢 Excellent	↓52% slower
PASS	select_count	Select	SELECT COUNT(*) on a 200-row table	100	1.3ms	1.2ms	1.5ms	3.2ms	1.0ms	3.3ms	781	128.1ms	🟢 Excellent	↓48% slower
PASS	select_order_by_limit	Select	SELECT with ORDER BY + LIMIT 10 on a 200-row table	100	1.5ms	1.5ms	1.8ms	3.2ms	1.3ms	3.6ms	650	154.0ms	🟢 Excellent	↓36% slower
PASS	single_update	Update	UPDATE a single row by filter condition	100	1000µs	890µs	1.2ms	4.2ms	689µs	4.3ms	1.0K	100.0ms	🟢 Excellent	↓84% slower
PASS	single_delete	Delete	DELETE a single row by filter condition	100	851µs	740µs	1.0ms	4.2ms	600µs	5.1ms	1.2K	85.1ms	🟢 Excellent	↓93% slower
PASS	concurrent_insert	Concurrent	N concurrent INSERT operations in parallel (N = concurrency setting)	100	2.2ms	1.3ms	11.5ms	12.9ms	943µs	13.3ms	456	219.5ms	🟢 Excellent	↓257% slower
PASS	concurrent_select	Concurrent	N concurrent SELECT operations in parallel (N = concurrency setting)	100	1.6ms	1.5ms	1.8ms	4.2ms	1.3ms	4.4ms	630	158.7ms	🟢 Excellent	↓72% slower
PASS	point_lookup	Select	SELECT by primary key from a 10K-row table (single row lookup)	100	693µs	658µs	848µs	924µs	522µs	2.2ms	1.4K	69.3ms	🟢 Excellent	↓59% slower
PASS	aggregate_query	Select	GROUP BY + SUM/AVG/COUNT on a 10K-row table (analytical query performance)	100	40.6ms	40.5ms	42.8ms	43.8ms	37.3ms	45.2ms	25	4.06s	🔴 Slow	↓7% slower
PASS	multi_table_join	Select	SELECT with subquery across two tables (200 customers, 1000 orders)	100	4.7ms	4.7ms	5.2ms	6.1ms	4.2ms	6.3ms	212	471.1ms	🟢 Excellent	↓18% slower
PASS	large_payload_insert	Insert	INSERT rows with ~4KB TEXT payloads (serialization + storage throughput)	100	2.2ms	2.0ms	3.0ms	3.2ms	1.6ms	3.3ms	454	220.3ms	🟢 Excellent	↓147% slower
PASS	wide_column_insert	Insert	INSERT into a 20-column table (wide schema overhead)	100	1.1ms	1.0ms	1.3ms	2.0ms	862µs	2.2ms	943	106.0ms	🟢 Excellent	↓66% slower
PASS	bulk_delete	Delete	DELETE 100 rows at once with a range filter (bulk deletion)	100	182.9ms	183.3ms	199.2ms	201.9ms	162.3ms	211.2ms	5	18.29s	🟡 Acceptable	↓4% slower
PASS	sequential_crud	DML	INSERT → UPDATE → SELECT → DELETE full DML lifecycle per iteration	100	3.0ms	2.7ms	5.8ms	6.1ms	2.3ms	6.3ms	334	299.6ms	🟢 Excellent	↓55% slower
PASS	alter_table	DDL	ALTER TABLE ADD COLUMN + DROP COLUMN (schema evolution latency)	100	3.2ms	2.9ms	6.0ms	6.5ms	2.2ms	6.5ms	314	318.6ms	🟢 Excellent	↓62% slower
PASS	concurrent_update	Concurrent	N concurrent UPDATE operations on the same table (write contention test)	100	4.1ms	3.5ms	10.9ms	11.1ms	2.0ms	11.6ms	243	411.0ms	🟢 Excellent	↓100% slower
PASS	concurrent_mixed_dml	Concurrent	Concurrent INSERT + UPDATE + DELETE on the same table (multi-op contention)	100	4.8ms	4.5ms	8.4ms	8.7ms	3.2ms	8.8ms	207	483.7ms	🟢 Excellent	↓27% slower
PASS	namespace_isolation	Concurrent	Concurrent queries across 5 different namespaces (isolation test)	100	4.3ms	4.1ms	8.8ms	9.1ms	2.8ms	9.2ms	231	433.5ms	🟢 Excellent	↓109% slower
PASS	subscribe_initial_load	Subscribe	Subscribe to a 1000-row user table and receive the full initial data batch	100	11.0ms	10.5ms	11.8ms	18.9ms	9.9ms	29.5ms	91	1.10s	🟢 Excellent	~ 10808µs prior
PASS	subscribe_change_latency	Subscribe	Latency from INSERT to subscriber receiving the change notification	100	5.02s	5.01s	5.02s	5.21s	5.00s	5.26s	0	502.03s	🔴 Slow	~ 5018104µs prior
PASS	reconnect_subscribe	Subscribe	Disconnect and re-subscribe to a user table (reconnection overhead)	100	11.9ms	12.0ms	12.7ms	12.9ms	10.4ms	13.0ms	84	1.19s	🟢 Excellent	↑27% faster
PASS	flushed_parquet_query	Storage	SELECT from a shared table with 20 flushed Parquet files (200K rows)	100	260.7ms	309.4ms	360.1ms	367.2ms	92.0ms	371.6ms	4	26.07s	🔴 Slow	—
PASS	concurrent_subscribers	Load	N WebSocket live-query subscribers receiving changes from concurrent writes	100	1.58s	1.57s	1.61s	1.64s	1.54s	1.65s	1	157.76s	🟢 Excellent	↓4% slower
PASS	concurrent_publishers	Load	N concurrent INSERTs into a topic-sourced table (measures publish overhead)	100	2.2ms	1.3ms	9.2ms	13.0ms	896µs	13.2ms	459	217.7ms	🟢 Excellent	↓187% slower
PASS	concurrent_consumers	Load	N concurrent topic CONSUME calls pulling messages in parallel	100	616µs	612µs	683µs	701µs	546µs	747µs	1.6K	61.6ms	🟢 Excellent	~ 628µs prior
PASS	sql_1k_concurrent	Load	1000 concurrent SQL SELECT queries at once (RPS degradation test)	100	429.8ms	318.8ms	911.4ms	2.80s	259.4ms	3.12s	2	42.98s	🟢 Excellent	↓64% slower
PASS	create_user	Load	CREATE USER (auth subsystem stress test)	100	1.9ms	1.7ms	2.7ms	3.9ms	1.4ms	5.9ms	533	187.6ms	🟢 Excellent	↓10% slower
PASS	drop_user	Load	DROP USER (auth subsystem teardown stress test)	100	551µs	502µs	813µs	1.1ms	397µs	1.3ms	1.8K	55.1ms	🟢 Excellent	↓5% slower
PASS	connection_storm	Load	N simultaneous login + SQL + cycles (connection setup overhead)	100	6.2ms	5.6ms	9.5ms	10.3ms	4.5ms	11.0ms	162	618.0ms	🟢 Excellent	↓29% slower
PASS	mixed_read_write	Load	50/50 concurrent reads + writes on same table (contention test)	100	3.8ms	3.5ms	7.8ms	8.5ms	2.1ms	8.6ms	261	383.1ms	🟢 Excellent	↓27% slower
PASS	wide_fanout_query	Load	N concurrent large-result-set SELECTs (serialization + memory pressure)	100	19.2ms	17.9ms	30.9ms	35.6ms	10.7ms	39.2ms	52	1.92s	🟢 Excellent	↓53% slower
PASS	subscriber_scale	Scale	Progressive subscriber scale test (up to --max-subscribers, default 100K)	1	10.36s	10.36s	10.36s	10.36s	10.36s	10.36s	0	10.36s	🟢 Excellent	—