Commit Latency vs Throughput
Commit Latency vs Throughput
Analysis of how commit latency scales with achieved throughput.
Values shown as mean ± standard deviation across seeds.
real_conflict_probability = 0.0
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.3 ± 0.0 | 1.9 ± 0.2 | 3.4 ± 0.5 | 2.1 |
| 0.5 ± 0.0 | 100.0 ± 0.0 | 0.7 ± 0.0 | 4.2 ± 0.3 | 8.1 ± 0.6 | 2.4 |
| 1.0 ± 0.0 | 100.0 ± 0.0 | 1.3 ± 0.0 | 8.2 ± 0.8 | 15.6 ± 1.1 | 2.7 |
| 2.0 ± 0.0 | 100.0 ± 0.0 | 2.6 ± 0.1 | 17.2 ± 0.6 | 32.8 ± 1.7 | 3.2 |
| 4.9 ± 0.1 | 100.0 ± 0.0 | 6.4 ± 0.2 | 43.2 ± 0.9 | 83.6 ± 3.4 | 4.3 |
| 9.8 ± 0.1 | 98.8 ± 0.1 | 13.6 ± 0.1 | 90.3 ± 1.1 | 170.9 ± 2.3 | 6.0 |
| 15.1 ± 0.0 | 76.4 ± 0.5 | 21.4 ± 0.2 | 132.4 ± 2.2 | 252.7 ± 3.6 | 7.9 |
| 17.6 ± 0.0 | 35.0 ± 0.1 | 24.2 ± 0.1 | 143.6 ± 2.3 | 274.7 ± 4.7 | 8.5 |
| 18.6 ± 0.0 | 18.0 ± 0.0 | 25.8 ± 0.1 | 145.4 ± 1.9 | 272.5 ± 2.6 | 8.7 |
real_conflict_probability = 0.1
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.4 ± 0.0 | 1.7 ± 0.2 | 2.8 ± 0.3 | 2.1 |
| 0.5 ± 0.0 | 100.0 ± 0.0 | 0.7 ± 0.0 | 4.5 ± 0.8 | 7.9 ± 0.8 | 2.4 |
| 1.0 ± 0.0 | 100.0 ± 0.0 | 1.4 ± 0.0 | 8.4 ± 0.8 | 15.7 ± 1.6 | 2.8 |
| 2.0 ± 0.0 | 100.0 ± 0.0 | 2.7 ± 0.0 | 17.4 ± 0.4 | 34.3 ± 1.3 | 3.3 |
| 4.9 ± 0.1 | 100.0 ± 0.0 | 6.6 ± 0.1 | 44.5 ± 1.8 | 87.1 ± 3.4 | 4.4 |
| 9.6 ± 0.1 | 97.7 ± 0.2 | 13.4 ± 0.2 | 89.3 ± 1.2 | 169.0 ± 3.0 | 6.1 |
| 14.6 ± 0.0 | 73.6 ± 0.2 | 20.9 ± 0.2 | 129.2 ± 2.5 | 240.9 ± 2.7 | 7.9 |
| 17.1 ± 0.0 | 34.2 ± 0.1 | 23.6 ± 0.2 | 140.2 ± 2.2 | 272.5 ± 3.7 | 8.4 |
| 18.3 ± 0.0 | 17.8 ± 0.0 | 25.3 ± 0.1 | 143.6 ± 0.4 | 272.8 ± 2.0 | 8.6 |
real_conflict_probability = 0.2
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.4 ± 0.0 | 1.8 ± 0.2 | 3.4 ± 0.5 | 2.1 |
| 0.5 ± 0.0 | 100.0 ± 0.0 | 0.8 ± 0.0 | 4.5 ± 0.2 | 8.4 ± 0.9 | 2.4 |
| 1.0 ± 0.0 | 100.0 ± 0.0 | 1.4 ± 0.1 | 8.6 ± 0.5 | 15.4 ± 1.1 | 2.8 |
| 2.0 ± 0.0 | 100.0 ± 0.0 | 2.7 ± 0.1 | 18.0 ± 0.8 | 34.5 ± 1.6 | 3.3 |
| 4.9 ± 0.0 | 99.9 ± 0.0 | 6.6 ± 0.1 | 44.3 ± 1.1 | 84.2 ± 3.5 | 4.5 |
| 9.5 ± 0.0 | 96.1 ± 0.2 | 13.4 ± 0.1 | 86.6 ± 1.1 | 165.4 ± 2.4 | 6.3 |
| 14.0 ± 0.0 | 70.4 ± 0.2 | 20.2 ± 0.1 | 123.6 ± 1.5 | 227.3 ± 1.6 | 7.8 |
| 16.7 ± 0.0 | 33.2 ± 0.2 | 23.0 ± 0.1 | 134.5 ± 1.6 | 265.0 ± 3.9 | 8.4 |
| 17.9 ± 0.0 | 17.4 ± 0.0 | 24.8 ± 0.2 | 143.1 ± 1.9 | 272.4 ± 4.4 | 8.6 |
real_conflict_probability = 0.3
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.5 ± 0.0 | 1.8 ± 0.1 | 3.4 ± 0.2 | 2.1 |
| 0.5 ± 0.0 | 100.0 ± 0.0 | 0.8 ± 0.1 | 4.5 ± 0.4 | 8.6 ± 0.3 | 2.4 |
| 1.0 ± 0.0 | 100.0 ± 0.0 | 1.5 ± 0.1 | 8.8 ± 0.4 | 15.8 ± 1.1 | 2.8 |
| 2.0 ± 0.0 | 100.0 ± 0.0 | 2.7 ± 0.1 | 17.6 ± 0.6 | 32.8 ± 1.6 | 3.4 |
| 4.9 ± 0.1 | 99.8 ± 0.0 | 6.7 ± 0.1 | 43.8 ± 0.9 | 85.0 ± 2.1 | 4.7 |
| 9.3 ± 0.1 | 93.8 ± 0.1 | 13.1 ± 0.2 | 84.8 ± 1.0 | 158.6 ± 5.3 | 6.4 |
| 13.3 ± 0.0 | 67.1 ± 0.2 | 19.2 ± 0.0 | 118.3 ± 1.7 | 225.4 ± 3.9 | 7.7 |
| 16.0 ± 0.0 | 32.0 ± 0.1 | 22.4 ± 0.0 | 130.0 ± 2.2 | 257.7 ± 6.3 | 8.3 |
| 17.4 ± 0.0 | 16.9 ± 0.0 | 24.4 ± 0.2 | 140.6 ± 1.4 | 270.4 ± 3.4 | 8.5 |
real_conflict_probability = 0.5
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.5 ± 0.0 | 2.0 ± 0.2 | 3.8 ± 0.5 | 2.1 |
| 0.5 ± 0.0 | 100.0 ± 0.0 | 0.9 ± 0.0 | 4.3 ± 0.4 | 8.1 ± 0.5 | 2.4 |
| 1.0 ± 0.0 | 100.0 ± 0.0 | 1.6 ± 0.0 | 9.0 ± 0.2 | 17.8 ± 1.3 | 2.9 |
| 2.0 ± 0.0 | 100.0 ± 0.0 | 2.9 ± 0.0 | 17.7 ± 0.6 | 33.5 ± 1.9 | 3.6 |
| 4.8 ± 0.0 | 98.3 ± 0.2 | 6.9 ± 0.1 | 43.1 ± 1.7 | 83.8 ± 3.6 | 5.2 |
| 8.5 ± 0.1 | 86.4 ± 0.3 | 12.1 ± 0.2 | 76.2 ± 1.2 | 145.1 ± 1.6 | 6.6 |
| 12.0 ± 0.0 | 60.3 ± 0.2 | 17.1 ± 0.1 | 103.9 ± 1.8 | 199.8 ± 5.0 | 7.6 |
| 14.6 ± 0.0 | 29.0 ± 0.1 | 21.0 ± 0.1 | 121.7 ± 1.7 | 232.9 ± 3.0 | 8.2 |
| 16.1 ± 0.0 | 15.7 ± 0.1 | 22.6 ± 0.1 | 131.8 ± 2.7 | 261.7 ± 5.9 | 8.4 |
real_conflict_probability = 0.7
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.5 ± 0.0 | 2.0 ± 0.2 | 3.3 ± 0.6 | 2.1 |
| 0.5 ± 0.0 | 100.0 ± 0.0 | 1.0 ± 0.0 | 4.6 ± 0.5 | 8.6 ± 1.1 | 2.5 |
| 1.0 ± 0.0 | 100.0 ± 0.0 | 1.7 ± 0.0 | 9.2 ± 0.5 | 17.6 ± 1.6 | 3.0 |
| 2.0 ± 0.0 | 99.9 ± 0.0 | 3.1 ± 0.1 | 17.5 ± 1.1 | 33.6 ± 2.4 | 3.8 |
| 4.6 ± 0.0 | 92.6 ± 0.2 | 6.8 ± 0.1 | 39.6 ± 0.5 | 76.1 ± 0.8 | 5.6 |
| 7.3 ± 0.0 | 74.2 ± 0.2 | 10.8 ± 0.1 | 64.7 ± 1.6 | 125.0 ± 4.1 | 6.7 |
| 10.3 ± 0.1 | 51.9 ± 0.3 | 14.9 ± 0.1 | 89.3 ± 0.8 | 174.9 ± 4.7 | 7.5 |
| 13.0 ± 0.0 | 26.0 ± 0.1 | 18.8 ± 0.1 | 108.3 ± 1.1 | 211.1 ± 8.2 | 8.2 |
| 14.3 ± 0.0 | 13.8 ± 0.0 | 20.7 ± 0.1 | 119.4 ± 2.4 | 221.9 ± 5.6 | 8.3 |
real_conflict_probability = 1.0
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.6 ± 0.0 | 2.2 ± 0.1 | 3.9 ± 0.3 | 2.1 |
| 0.5 ± 0.0 | 100.0 ± 0.0 | 1.1 ± 0.0 | 4.6 ± 0.2 | 7.7 ± 0.6 | 2.5 |
| 1.0 ± 0.0 | 100.0 ± 0.0 | 1.9 ± 0.0 | 9.2 ± 0.2 | 17.6 ± 0.6 | 3.2 |
| 1.9 ± 0.0 | 98.7 ± 0.2 | 3.4 ± 0.1 | 17.5 ± 0.9 | 32.9 ± 1.6 | 4.4 |
| 3.2 ± 0.0 | 63.9 ± 0.4 | 5.7 ± 0.0 | 27.7 ± 0.6 | 52.3 ± 0.7 | 6.6 |
| 3.4 ± 0.0 | 34.4 ± 0.2 | 6.0 ± 0.0 | 29.0 ± 0.3 | 58.1 ± 2.0 | 6.9 |
| 3.5 ± 0.0 | 17.9 ± 0.1 | 6.1 ± 0.1 | 30.1 ± 1.0 | 58.7 ± 1.4 | 6.9 |
| 3.7 ± 0.0 | 7.3 ± 0.0 | 6.3 ± 0.1 | 31.5 ± 0.8 | 60.6 ± 1.6 | 7.0 |
| 3.8 ± 0.0 | 3.6 ± 0.0 | 6.5 ± 0.0 | 32.5 ± 1.1 | 62.1 ± 1.4 | 7.1 |
Notes
- Latencies reported in seconds (converted from milliseconds)
- Values shown as mean ± standard deviation across multiple seeds
- Throughput = commits per second during steady-state window
- Success rate = percentage of transactions that committed successfully
- Mean retries = average number of retry attempts per committed transaction
