m8g.medium
General purpose · AWS Graviton4 (Arm)
Compute
- vCPU
- 1
- Memory
- 4 GiB
- Mem / vCPU
- 4 : 1
Platform
- Family
- m8g
- Architecture
- arm64
- Performance
- Fixed
Cost & region
- On-demand
- $0.0449/hr~$33/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 (~3 GB of 4 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 | 16,670±0.4% | 1.92±0.4% | 3.44±0.5% | 4.28±0.6% | 249,555 | 16,670 | 4,167 |
| 1 GB | gp3-max | 16,487±0.2% | 1.94±0.2% | 3.53±0.2% | 4.31±0.2% | 80,903 | 16,487 | 4,122 |
| 10 GB | gp3-max | 12,618±0.3% | 2.54±0.3% | 3.74±0.3% | 5.01±0.8% | 61,919 | 12,618 | 3,155 |
| 10 GB | gp3-baseline | 3,537±0.1% | 9.05±0.1% | 13.81±0.2% | 16.48±0.3% | 52,952 | 3,537 | 884 |
| 50 GB | gp3-max | 10,752±0.2% | 2.98±0.2% | 4.04±0.2% | 6.04±0.4% | 52,762 | 10,752 | 2,688 |
| 50 GB | gp3-baseline | 2,028±0.1% | 15.78±0.1% | 21.58±0.3% | 23.85±0.4% | 30,357 | 2,028 | 507 |