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 | 0.69 ± 0.01 | 1.57 ± 0.18 | 4.32 ± 0.86 | 1.1 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.69 ± 0.00 | 3.74 ± 0.77 | 11.06 ± 0.97 | 1.2 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.70 ± 0.00 | 5.70 ± 1.21 | 23.74 ± 3.32 | 1.4 |
| 1.6 ± 0.0 | 99.2 ± 0.2 | 0.95 ± 0.07 | 10.74 ± 0.80 | 44.52 ± 1.20 | 2.2 |
| 2.0 ± 0.0 | 96.8 ± 0.4 | 1.08 ± 0.01 | 11.97 ± 1.32 | 52.76 ± 3.23 | 2.7 |
| 2.4 ± 0.0 | 89.4 ± 0.6 | 1.41 ± 0.03 | 9.46 ± 0.93 | 56.17 ± 3.79 | 3.4 |
| 3.0 ± 0.0 | 72.5 ± 0.5 | 1.80 ± 0.02 | 5.65 ± 1.00 | 48.10 ± 7.11 | 4.2 |
| 3.5 ± 0.0 | 43.3 ± 0.3 | 2.20 ± 0.01 | 4.26 ± 0.02 | 9.80 ± 4.45 | 5.0 |
| 3.8 ± 0.0 | 23.0 ± 0.1 | 2.27 ± 0.01 | 4.24 ± 0.01 | 4.41 ± 0.02 | 5.3 |
| 4.0 ± 0.0 | 9.7 ± 0.0 | 2.32 ± 0.03 | 4.23 ± 0.01 | 4.39 ± 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 | 0.68 ± 0.00 | 1.09 ± 0.11 | 2.53 ± 1.03 | 1.1 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.69 ± 0.00 | 1.71 ± 0.16 | 5.47 ± 0.56 | 1.2 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.70 ± 0.00 | 3.39 ± 0.14 | 12.63 ± 1.30 | 1.4 |
| 1.7 ± 0.0 | 99.9 ± 0.0 | 0.87 ± 0.01 | 5.71 ± 0.33 | 23.75 ± 1.21 | 1.9 |
| 2.1 ± 0.0 | 99.5 ± 0.1 | 0.94 ± 0.00 | 7.43 ± 0.43 | 27.37 ± 1.29 | 2.3 |
| 2.7 ± 0.0 | 96.8 ± 0.3 | 1.13 ± 0.01 | 8.50 ± 0.51 | 36.61 ± 1.23 | 3.0 |
| 3.4 ± 0.0 | 83.8 ± 0.2 | 1.57 ± 0.03 | 7.72 ± 0.73 | 41.15 ± 3.94 | 4.1 |
| 4.1 ± 0.0 | 50.0 ± 0.1 | 2.07 ± 0.02 | 4.24 ± 0.08 | 34.26 ± 3.39 | 5.1 |
| 4.4 ± 0.0 | 26.7 ± 0.2 | 2.11 ± 0.02 | 4.10 ± 0.01 | 27.50 ± 2.60 | 5.2 |
| 4.7 ± 0.0 | 11.4 ± 0.1 | 2.14 ± 0.01 | 4.08 ± 0.01 | 17.41 ± 1.89 | 5.3 |
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 | 0.68 ± 0.00 | 0.95 ± 0.02 | 1.48 ± 0.29 | 1.1 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.69 ± 0.00 | 1.25 ± 0.08 | 3.06 ± 0.27 | 1.1 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.70 ± 0.00 | 1.79 ± 0.21 | 5.08 ± 0.74 | 1.3 |
| 1.6 ± 0.0 | 100.0 ± 0.0 | 0.82 ± 0.03 | 2.91 ± 0.24 | 9.34 ± 0.90 | 1.7 |
| 2.1 ± 0.0 | 100.0 ± 0.0 | 0.90 ± 0.01 | 3.53 ± 0.26 | 12.03 ± 1.18 | 2.0 |
| 2.7 ± 0.0 | 99.7 ± 0.0 | 0.96 ± 0.01 | 4.19 ± 0.25 | 14.45 ± 0.74 | 2.5 |
| 3.9 ± 0.0 | 96.6 ± 0.3 | 1.28 ± 0.01 | 5.96 ± 0.53 | 21.92 ± 1.65 | 3.5 |
| 6.0 ± 0.0 | 73.3 ± 0.4 | 1.87 ± 0.01 | 7.56 ± 0.45 | 31.10 ± 2.23 | 5.1 |
| 7.1 ± 0.0 | 43.4 ± 0.1 | 2.20 ± 0.01 | 6.87 ± 0.34 | 33.94 ± 1.75 | 5.8 |
| 7.7 ± 0.0 | 18.8 ± 0.1 | 2.35 ± 0.01 | 5.88 ± 0.30 | 32.40 ± 2.18 | 6.1 |
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 | 0.67 ± 0.01 | 0.89 ± 0.05 | 1.16 ± 0.13 | 1.0 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.68 ± 0.00 | 1.04 ± 0.02 | 1.70 ± 0.06 | 1.1 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.70 ± 0.01 | 1.37 ± 0.09 | 2.95 ± 0.35 | 1.3 |
| 1.6 ± 0.0 | 100.0 ± 0.0 | 0.80 ± 0.02 | 2.02 ± 0.12 | 5.53 ± 0.43 | 1.7 |
| 2.1 ± 0.0 | 100.0 ± 0.0 | 0.89 ± 0.00 | 2.32 ± 0.04 | 6.49 ± 0.45 | 1.9 |
| 2.7 ± 0.0 | 99.9 ± 0.0 | 0.94 ± 0.00 | 2.89 ± 0.16 | 8.72 ± 0.90 | 2.3 |
| 4.0 ± 0.0 | 98.3 ± 0.1 | 1.16 ± 0.01 | 3.60 ± 0.22 | 12.14 ± 0.59 | 3.2 |
| 6.5 ± 0.0 | 78.8 ± 0.5 | 1.67 ± 0.01 | 5.44 ± 0.16 | 19.05 ± 0.52 | 4.9 |
| 7.6 ± 0.0 | 46.1 ± 0.1 | 2.02 ± 0.00 | 5.37 ± 0.19 | 20.16 ± 0.98 | 5.8 |
| 8.0 ± 0.0 | 19.4 ± 0.0 | 2.12 ± 0.02 | 5.61 ± 0.27 | 21.16 ± 0.67 | 6.0 |
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 | 0.65 ± 0.00 | 0.81 ± 0.02 | 0.97 ± 0.03 | 1.0 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.68 ± 0.00 | 0.98 ± 0.03 | 1.27 ± 0.12 | 1.1 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.70 ± 0.00 | 1.19 ± 0.02 | 2.05 ± 0.18 | 1.3 |
| 1.6 ± 0.0 | 100.0 ± 0.0 | 0.78 ± 0.02 | 1.60 ± 0.02 | 3.19 ± 0.24 | 1.6 |
| 2.1 ± 0.0 | 100.0 ± 0.0 | 0.89 ± 0.00 | 1.78 ± 0.04 | 3.88 ± 0.36 | 1.9 |
| 2.7 ± 0.0 | 99.8 ± 0.1 | 0.93 ± 0.00 | 2.24 ± 0.06 | 4.92 ± 0.29 | 2.2 |
| 4.0 ± 0.0 | 98.4 ± 0.1 | 1.12 ± 0.00 | 2.86 ± 0.05 | 6.86 ± 0.30 | 3.1 |
| 6.6 ± 0.0 | 80.0 ± 0.1 | 1.58 ± 0.01 | 3.48 ± 0.05 | 10.08 ± 0.32 | 4.7 |
| 7.7 ± 0.0 | 46.9 ± 0.2 | 1.93 ± 0.01 | 3.86 ± 0.13 | 11.51 ± 0.49 | 5.8 |
| 8.0 ± 0.0 | 19.5 ± 0.1 | 2.05 ± 0.01 | 3.94 ± 0.10 | 11.53 ± 0.43 | 6.0 |
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 | 0.59 ± 0.02 | 0.78 ± 0.02 | 0.91 ± 0.07 | 1.0 |
| 0.4 ± 0.0 | 100.0 ± 0.0 | 0.66 ± 0.01 | 0.93 ± 0.03 | 1.13 ± 0.05 | 1.1 |
| 0.8 ± 0.0 | 100.0 ± 0.0 | 0.68 ± 0.00 | 1.05 ± 0.02 | 1.37 ± 0.05 | 1.2 |
| 1.7 ± 0.0 | 100.0 ± 0.0 | 0.75 ± 0.01 | 1.36 ± 0.02 | 1.92 ± 0.04 | 1.6 |
| 2.0 ± 0.0 | 100.0 ± 0.0 | 0.86 ± 0.00 | 1.48 ± 0.02 | 2.21 ± 0.08 | 1.8 |
| 2.7 ± 0.0 | 99.9 ± 0.1 | 0.92 ± 0.00 | 1.82 ± 0.03 | 2.73 ± 0.07 | 2.2 |
| 4.0 ± 0.0 | 98.5 ± 0.2 | 1.07 ± 0.01 | 2.39 ± 0.02 | 3.55 ± 0.05 | 3.0 |
| 6.6 ± 0.0 | 80.3 ± 0.3 | 1.45 ± 0.01 | 2.94 ± 0.01 | 5.38 ± 0.21 | 4.6 |
| 7.7 ± 0.0 | 47.2 ± 0.3 | 1.89 ± 0.01 | 3.06 ± 0.01 | 6.10 ± 0.23 | 5.8 |
| 8.0 ± 0.0 | 19.6 ± 0.1 | 1.97 ± 0.01 | 3.10 ± 0.02 | 6.01 ± 0.18 | 6.0 |
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
