Performance History

Track Kapper's performance improvements and competitive positioning across tested versions.

Version Performance Timeline

Kapper 1.6.1 (Latest)

Release Date: Aug 2025

Major Performance Improvements:

• Overall improvement: 7.5% average performance gain across 50 benchmark comparisons

Top Performance Gains:

• findById vs Ktorm: 81.9% improvement (1181.6 → 213.9 μs/op)
• find100 operations: 47.9% improvement (401.8 → 209.5 μs/op)
• findById operations: 15.9% improvement (71.4 → 60.1 μs/op)

Competitive Analysis:

Library Comparisons (1000-row operations):
┌──────────────────┬─────────────┬─────────────┬─────────────┐
│ Operation        │ vs JDBC     │ vs Hibernate│ vs Ktorm    │
├──────────────────┼─────────────┼─────────────┼─────────────┤
│ findById (SQLite)│ 69.7% slower│ 45% faster  │ 88.3% faster│
│ find100 (SQLite) │ 61.3% slower│ 23% faster  │ 26% faster  │
│ insert (SQLite)  │ 12% slower  │ 67% faster  │ 89.8% faster│
│ update (SQLite)  │ 15% slower  │ 74% faster  │ 94.1% faster│
│ simpleJoin       │ 40% slower  │ 34% faster  │ 88.6% faster│
└──────────────────┴─────────────┴─────────────┴─────────────┘

Kapper 1.5.0

Release Date: June 2025

Benchmark Results:

Main Operations (μs/op):
┌──────────────────┬─────────────┬─────────────┬─────────────┬─────────────┐
│ Operation        │ JDBC        │ KAPPER      │ HIBERNATE   │ KTORM       │
├──────────────────┼─────────────┼─────────────┼─────────────┼─────────────┤
│ find100 (SQLite) │ 154.6 ± 2.1 │ 401.8 ± 8.2 │ 288.9 ± 5.1 │ 389.1 ± 7.8 │
│ findById (SQLite)│ 10.9 ± 0.3  │ 71.4 ± 1.2  │ 23.7 ± 0.8  │ 1181.6 ± 45 │
│ insert (SQLite)  │ 6.2 ± 0.2   │ 7.1 ± 0.3   │ 23.1 ± 1.1  │ 162.0 ± 3.2 │
│ update (SQLite)  │ 6.7 ± 0.2   │ 7.7 ± 0.3   │ 31.8 ± 1.4  │ 149.5 ± 2.9 │
│ simpleJoin       │ 18.6 ± 0.7  │ 29.2 ± 1.1  │ 39.1 ± 1.8  │ 257.1 ± 5.4 │
└──────────────────┴─────────────┴─────────────┴─────────────┴─────────────┘

Kapper 1.3.0

Release Date: March 2025

Available legacy data shows comprehensive performance metrics with error bars and percentile data, used for visualization testing and trend analysis.

Benchmark Methodology

Environment

• Platform: Oracle Cloud ARM instances (4 OCPUs, 24GB RAM)
• JVM: GraalVM JDK 21 with optimized JIT compilation
• Framework: JMH (Java Microbenchmark Harness)
• Databases: SQLite 3.50, PostgreSQL 14 (Testcontainers)

Measurements

• Warmup: 5 iterations, 2 seconds each
• Execution: 5 iterations, 5 seconds each
• Mode: Average time per operation (AVGT)
• Statistics: Score ± error with percentile data

Data Integrity

• All benchmarks run on identical infrastructure
• Fresh database state for each test
• Statistical validation with confidence intervals
• Version-controlled benchmark code

Next Steps

• Performance Comparisons - Detailed competitive analysis