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.
num_tables = 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 | 0.47 ± 0.00 | 1.18 ± 0.30 | 4.04 ± 0.61 | 1.1 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.48 ± 0.00 | 3.10 ± 0.58 | 12.70 ± 2.44 | 1.2 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.49 ± 0.00 | 5.24 ± 0.89 | 23.10 ± 1.50 | 1.3 |
| 1.6 ± 0.0 | 100.0 ± 0.0 | 0.51 ± 0.01 | 9.85 ± 2.74 | 42.00 ± 5.25 | 1.7 |
| 2.0 ± 0.0 | 99.6 ± 0.1 | 0.56 ± 0.01 | 11.75 ± 1.43 | 54.53 ± 4.38 | 2.0 |
| 2.7 ± 0.0 | 97.6 ± 0.5 | 0.72 ± 0.01 | 13.36 ± 0.93 | 68.68 ± 5.80 | 2.6 |
| 3.5 ± 0.0 | 86.0 ± 0.4 | 0.97 ± 0.01 | 7.38 ± 0.82 | 63.56 ± 7.87 | 3.6 |
| 4.5 ± 0.0 | 54.7 ± 0.2 | 1.30 ± 0.01 | 2.82 ± 0.01 | 27.62 ± 4.44 | 4.6 |
| 5.1 ± 0.0 | 31.1 ± 0.1 | 1.48 ± 0.02 | 2.77 ± 0.01 | 2.96 ± 0.02 | 5.1 |
| 5.6 ± 0.0 | 13.6 ± 0.0 | 1.54 ± 0.02 | 2.76 ± 0.00 | 2.92 ± 0.01 | 5.4 |
num_tables = 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 | 0.48 ± 0.00 | 0.86 ± 0.09 | 2.81 ± 0.72 | 1.0 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.48 ± 0.00 | 1.57 ± 0.40 | 5.21 ± 0.96 | 1.1 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.48 ± 0.00 | 2.61 ± 0.35 | 11.22 ± 0.95 | 1.3 |
| 1.7 ± 0.0 | 100.0 ± 0.0 | 0.51 ± 0.00 | 5.70 ± 0.52 | 24.30 ± 3.11 | 1.6 |
| 2.1 ± 0.0 | 100.0 ± 0.0 | 0.53 ± 0.00 | 5.84 ± 0.49 | 26.18 ± 1.48 | 1.7 |
| 2.7 ± 0.0 | 99.9 ± 0.0 | 0.57 ± 0.00 | 7.88 ± 0.74 | 36.71 ± 1.89 | 2.1 |
| 4.0 ± 0.0 | 98.5 ± 0.1 | 0.74 ± 0.01 | 10.37 ± 1.24 | 51.55 ± 4.60 | 2.9 |
| 6.1 ± 0.0 | 74.1 ± 0.2 | 1.23 ± 0.01 | 3.46 ± 0.98 | 45.86 ± 4.32 | 5.0 |
| 6.8 ± 0.0 | 41.4 ± 0.1 | 1.45 ± 0.01 | 2.59 ± 0.01 | 18.02 ± 2.11 | 5.7 |
| 7.2 ± 0.0 | 17.6 ± 0.1 | 1.49 ± 0.01 | 2.58 ± 0.01 | 2.94 ± 0.07 | 5.8 |
num_tables = 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 | 0.47 ± 0.00 | 0.75 ± 0.07 | 1.28 ± 0.17 | 1.0 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.48 ± 0.00 | 0.92 ± 0.07 | 2.59 ± 0.60 | 1.1 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.48 ± 0.00 | 1.49 ± 0.24 | 4.90 ± 0.43 | 1.2 |
| 1.6 ± 0.0 | 100.0 ± 0.0 | 0.51 ± 0.00 | 2.42 ± 0.15 | 9.19 ± 0.21 | 1.5 |
| 2.1 ± 0.0 | 100.0 ± 0.0 | 0.52 ± 0.00 | 2.94 ± 0.23 | 11.98 ± 0.88 | 1.6 |
| 2.7 ± 0.0 | 100.0 ± 0.0 | 0.55 ± 0.00 | 3.57 ± 0.46 | 14.62 ± 1.42 | 1.8 |
| 4.1 ± 0.0 | 100.0 ± 0.0 | 0.59 ± 0.00 | 5.23 ± 0.34 | 22.05 ± 1.20 | 2.2 |
| 8.0 ± 0.0 | 97.3 ± 0.1 | 0.82 ± 0.00 | 9.21 ± 0.54 | 43.34 ± 2.30 | 3.7 |
| 12.3 ± 0.1 | 75.0 ± 0.3 | 1.19 ± 0.00 | 7.84 ± 0.20 | 52.81 ± 3.33 | 5.3 |
| 15.1 ± 0.0 | 36.8 ± 0.0 | 1.50 ± 0.00 | 2.48 ± 0.01 | 28.54 ± 2.33 | 6.2 |
num_tables = 10
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.46 ± 0.00 | 0.60 ± 0.03 | 0.88 ± 0.22 | 1.0 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.47 ± 0.00 | 0.73 ± 0.02 | 1.50 ± 0.20 | 1.1 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.48 ± 0.00 | 0.98 ± 0.07 | 2.71 ± 0.34 | 1.2 |
| 1.7 ± 0.0 | 100.0 ± 0.0 | 0.50 ± 0.00 | 1.50 ± 0.12 | 4.79 ± 0.18 | 1.4 |
| 2.1 ± 0.0 | 100.0 ± 0.0 | 0.52 ± 0.00 | 1.67 ± 0.16 | 5.66 ± 0.45 | 1.5 |
| 2.7 ± 0.0 | 100.0 ± 0.0 | 0.54 ± 0.00 | 2.07 ± 0.06 | 8.15 ± 0.41 | 1.7 |
| 4.1 ± 0.0 | 100.0 ± 0.0 | 0.57 ± 0.00 | 3.16 ± 0.16 | 11.74 ± 0.93 | 2.1 |
| 8.2 ± 0.0 | 99.2 ± 0.0 | 0.68 ± 0.01 | 5.37 ± 0.25 | 22.00 ± 0.46 | 3.2 |
| 14.0 ± 0.0 | 85.3 ± 0.1 | 0.95 ± 0.00 | 7.71 ± 0.22 | 36.01 ± 1.32 | 4.7 |
| 18.0 ± 0.0 | 43.9 ± 0.2 | 1.18 ± 0.00 | 7.26 ± 0.26 | 40.11 ± 1.07 | 5.9 |
num_tables = 20
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.44 ± 0.01 | 0.58 ± 0.01 | 0.77 ± 0.05 | 1.0 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.46 ± 0.00 | 0.65 ± 0.01 | 0.99 ± 0.05 | 1.1 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.48 ± 0.00 | 0.75 ± 0.02 | 1.62 ± 0.14 | 1.2 |
| 1.6 ± 0.0 | 100.0 ± 0.0 | 0.50 ± 0.00 | 0.99 ± 0.08 | 2.78 ± 0.31 | 1.4 |
| 2.1 ± 0.0 | 100.0 ± 0.0 | 0.52 ± 0.00 | 1.11 ± 0.05 | 3.38 ± 0.20 | 1.5 |
| 2.7 ± 0.0 | 100.0 ± 0.0 | 0.54 ± 0.00 | 1.37 ± 0.07 | 4.24 ± 0.42 | 1.7 |
| 4.1 ± 0.0 | 100.0 ± 0.0 | 0.57 ± 0.00 | 1.69 ± 0.11 | 6.13 ± 0.38 | 2.0 |
| 8.2 ± 0.0 | 99.5 ± 0.1 | 0.64 ± 0.00 | 3.06 ± 0.08 | 11.59 ± 0.40 | 2.9 |
| 14.3 ± 0.0 | 87.1 ± 0.2 | 0.84 ± 0.00 | 4.45 ± 0.21 | 19.12 ± 1.21 | 4.5 |
| 18.4 ± 0.0 | 45.0 ± 0.1 | 1.06 ± 0.00 | 5.09 ± 0.13 | 22.79 ± 0.62 | 5.8 |
num_tables = 50
| Throughput (c/s) | Success Rate (%) | P50 Latency (s) | P95 Latency (s) | P99 Latency (s) | Mean Retries |
|---|
| 0.2 ± 0.0 | 100.0 ± 0.0 | 0.36 ± 0.01 | 0.54 ± 0.01 | 0.62 ± 0.01 | 1.0 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.45 ± 0.00 | 0.62 ± 0.01 | 0.80 ± 0.04 | 1.1 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.47 ± 0.00 | 0.66 ± 0.01 | 0.94 ± 0.05 | 1.2 |
| 1.6 ± 0.0 | 100.0 ± 0.0 | 0.49 ± 0.00 | 0.76 ± 0.01 | 1.45 ± 0.04 | 1.4 |
| 2.0 ± 0.0 | 100.0 ± 0.0 | 0.51 ± 0.00 | 0.83 ± 0.02 | 1.74 ± 0.12 | 1.5 |
| 2.7 ± 0.0 | 100.0 ± 0.0 | 0.53 ± 0.00 | 0.91 ± 0.04 | 2.11 ± 0.12 | 1.6 |
| 4.1 ± 0.0 | 100.0 ± 0.0 | 0.56 ± 0.00 | 1.05 ± 0.04 | 2.71 ± 0.19 | 1.9 |
| 8.2 ± 0.0 | 99.4 ± 0.1 | 0.63 ± 0.00 | 1.58 ± 0.05 | 5.04 ± 0.35 | 2.8 |
| 14.3 ± 0.0 | 87.5 ± 0.2 | 0.78 ± 0.00 | 2.36 ± 0.09 | 8.33 ± 0.33 | 4.3 |
| 18.6 ± 0.1 | 45.3 ± 0.1 | 1.02 ± 0.00 | 2.80 ± 0.07 | 10.40 ± 0.22 | 5.9 |
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
