r8g.large
Memory optimized · AWS Graviton4 (Arm)
Compute
- vCPU
- 2
- Memory
- 16 GiB
- Mem / vCPU
- 8 : 1
Platform
- Family
- r8g
- Architecture
- arm64
- Performance
- Fixed
Cost & region
- On-demand
- $0.1045/hr~$76/mo
- Region
- us-east-1
Chart
Change the metric and disk to see how the numbers shift as the dataset grows — and where it stops fitting in memory.
Open markers sit past the RAM limit (~12 GB of 16 GiB): the dataset no longer fits in cache, so the working set spills to disk and throughput drops.
All results
6 of 6 rows
| 1 GB | gp3-baseline | 38,531±1.6% | 0.83±1.6% | 2.35±0.6% | 3.06±0.4% | 304,746 | 19,266 | 2,408 |
| 10 GB | gp3-baseline | 34,304±2.2% | 0.93±2.2% | 2.46±1.2% | 3.22±1.4% | 271,315 | 17,152 | 2,144 |
| 1 GB | gp3-max | 39,039±0.6% | 0.82±0.6% | 2.35±0.1% | 3.06±0.1% | 148,198 | 19,519 | 2,440 |
| 10 GB | gp3-max | 34,010±2.2% | 0.94±2.2% | 2.48±1.1% | 3.23±1.0% | 129,106 | 17,005 | 2,126 |
| 50 GB | gp3-max | 18,789±0.2% | 1.70±0.2% | 2.69±0.2% | 3.23±0.3% | 71,326 | 9,395 | 1,174 |
| 50 GB | gp3-baseline | 3,310±0.5% | 9.67±0.5% | 15.07±0.1% | 17.91±0.7% | 26,183 | 1,655 | 207 |