LMDB Library Performance Analysis

⚠️ SMOKETEST RESULTS

This report was generated from a smoketest run and should NOT be used for performance comparisons or production decisions. Smoketest results have:

• No warmup iterations
• Single iteration
• Minimal entry counts
• Short runtime

For valid performance results, run the benchmark script with benchmark mode instead.

Results | Trend

Performance Analysis

The following tables show each benchmark ranked by latency (smaller is better), with percentage difference from the fastest LMDB library build. The latest tag (LMDB_0.9.33) is highlighted in bold.

Read by Key

Rank	LMDB Tag	ms/op	vs Fastest
1	LMDB_0.9.19	0.056	baseline
2	LMDB_0.9.24	0.056	+1.0%
3	LMDB_0.9.23	0.057	+1.5%
4	LMDB_0.9.21	0.057	+1.5%
5	LMDB_0.9.30	0.057	+1.6%
6	LMDB_0.9.31	0.057	+1.7%
7	LMDB_0.9.22	0.057	+2.6%
8	LMDB_0.9.33	0.057	+2.9%
9	LMDB_0.9.28	0.058	+3.1%
10	LMDB_0.9.27	0.058	+3.5%
11	LMDB_0.9.29	0.058	+3.6%
12	LMDB_0.9.20	0.058	+4.5%
13	LMDB_0.9.18	0.062	+10.8%
14	LMDB_0.9.17	0.064	+14.5%

Write Entry

Rank	LMDB Tag	ms/op	vs Fastest
1	LMDB_0.9.19	0.149	baseline
2	LMDB_0.9.18	0.153	+2.7%
3	LMDB_0.9.17	0.154	+3.8%
4	LMDB_0.9.33	0.158	+6.4%
5	LMDB_0.9.30	0.159	+7.0%
6	LMDB_0.9.31	0.159	+7.1%
7	LMDB_0.9.29	0.159	+7.2%
8	LMDB_0.9.28	0.161	+8.0%
9	LMDB_0.9.21	0.161	+8.2%
10	LMDB_0.9.24	0.161	+8.3%
11	LMDB_0.9.22	0.162	+9.1%
12	LMDB_0.9.23	0.163	+9.8%
13	LMDB_0.9.27	0.164	+10.6%
14	LMDB_0.9.20	0.166	+12.0%

Calculate xxHash64

Rank	LMDB Tag	ms/op	vs Fastest
1	LMDB_0.9.31	0.058	baseline
2	LMDB_0.9.29	0.058	+0.1%
3	LMDB_0.9.22	0.058	+0.4%
4	LMDB_0.9.24	0.058	+0.9%
5	LMDB_0.9.23	0.058	+1.3%
6	LMDB_0.9.19	0.058	+1.4%
7	LMDB_0.9.18	0.059	+1.7%
8	LMDB_0.9.17	0.059	+1.8%
9	LMDB_0.9.30	0.059	+2.0%
10	LMDB_0.9.28	0.059	+2.1%
11	LMDB_0.9.20	0.059	+2.4%
12	LMDB_0.9.21	0.059	+2.4%
13	LMDB_0.9.33	0.059	+2.4%
14	LMDB_0.9.27	0.060	+4.3%

Iterate Sequentially

Rank	LMDB Tag	ms/op	vs Fastest
1	LMDB_0.9.29	0.027	baseline
2	LMDB_0.9.33	0.027	+0.2%
3	LMDB_0.9.21	0.027	+0.3%
4	LMDB_0.9.19	0.027	+0.6%
5	LMDB_0.9.20	0.027	+0.8%
6	LMDB_0.9.30	0.027	+1.0%
7	LMDB_0.9.31	0.027	+1.3%
8	LMDB_0.9.23	0.027	+1.3%
9	LMDB_0.9.24	0.027	+1.3%
10	LMDB_0.9.18	0.027	+2.1%
11	LMDB_0.9.17	0.027	+2.6%
12	LMDB_0.9.27	0.027	+2.6%
13	LMDB_0.9.22	0.028	+4.2%
14	LMDB_0.9.28	0.028	+4.9%

Iterate Reverse

Rank	LMDB Tag	ms/op	vs Fastest
1	LMDB_0.9.27	0.026	baseline
2	LMDB_0.9.21	0.026	+0.1%
3	LMDB_0.9.19	0.026	+0.5%
4	LMDB_0.9.29	0.026	+0.6%
5	LMDB_0.9.22	0.026	+0.7%
6	LMDB_0.9.28	0.026	+1.0%
7	LMDB_0.9.24	0.026	+1.2%
8	LMDB_0.9.33	0.026	+1.4%
9	LMDB_0.9.20	0.027	+1.5%
10	LMDB_0.9.30	0.027	+2.2%
11	LMDB_0.9.31	0.027	+2.5%
12	LMDB_0.9.23	0.027	+2.9%
13	LMDB_0.9.18	0.027	+4.5%
14	LMDB_0.9.17	0.028	+5.4%

Calculate CRC32

Rank	LMDB Tag	ms/op	vs Fastest
1	LMDB_0.9.24	0.076	baseline
2	LMDB_0.9.17	0.076	+0.0%
3	LMDB_0.9.22	0.076	+0.0%
4	LMDB_0.9.19	0.076	+0.0%
5	LMDB_0.9.29	0.076	+0.2%
6	LMDB_0.9.30	0.076	+0.3%
7	LMDB_0.9.33	0.076	+0.4%
8	LMDB_0.9.23	0.076	+0.5%
9	LMDB_0.9.20	0.076	+0.5%
10	LMDB_0.9.18	0.077	+0.6%
11	LMDB_0.9.28	0.077	+1.1%
12	LMDB_0.9.31	0.077	+1.4%
13	LMDB_0.9.21	0.078	+2.0%
14	LMDB_0.9.27	0.078	+2.1%

Test Configuration

The benchmark was executed on 2025-12-26 using LmdbJava Benchmarks.

Each test run uses -Dlmdbjava.native.lib to ensure use of a specific LMDB library built from the identified LMDB tag. This isolates LmdbJava wrapper code during each benchmark, ensuring the only variation is the actual LMDB library.

All tests use the LmdbJava Agrona implementation with the following configuration:

Test Environment

CPU	Intel(R) Xeon(R) Platinum 8370C CPU @ 2.80GHz (4 cores)
RAM	16 GiB
OS	Linux 6.11.0-1018-azure (x86_64)
Java	25.0.1
LmdbJava	master#83c3a62

Benchmark Configuration

• Implementation: LmdbJava Agrona only
• Test Profile: Run 4 sequential integer configuration (100-byte values)
• Key Type: Sequential 32-bit integers
• Value Size: 100 bytes
• Access Pattern: Sequential
• Benchmarks: All 6 operations (readCrc, readKey, readRev, readSeq, readXxh64, write)