t4g.large
Burstable general purpose · AWS Graviton2 (Arm)
PostgreSQL benchmark results for the t4g.large AWS EC2 instance with 2 vCPU, 8 GiB RAM, and arm64 architecture.
Compute
- vCPU
- 2
- Memory
- 8 GiB
- Mem / vCPU
- 4 : 1
Platform
- Family
- t4g
- Architecture
- arm64
- Performance
- Burstable
Cost & region
- On-demand
- $0.0672/hr~$49/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 (~6 GB of 8 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 | 5,157±0.3% | 57,870 | 6.20±0.3% | 56.39±0.1% | 62.81±0.1% | 2,579 | 645 |
| 10 GB | gp3-baseline | 3,924±6.1% | 44,026 | 8.18±5.8% | 58.16±1.4% | 61.86±0.8% | 1,962 | 490 |
| 50 GB | gp3-baseline | 2,725±0.2% | 30,575 | 11.74±0.2% | 17.48±0.6% | 20.36±1.2% | 1,362 | 341 |
| 1 GB | gp3-max | 5,028±0.3% | 22,238 | 6.36±0.3% | 57.73±0.0% | 60.23±0.1% | 2,514 | 629 |
| 10 GB | gp3-max | 3,759±1.8% | 16,623 | 8.52±1.8% | 58.85±0.1% | 61.28±0.1% | 1,879 | 470 |
| 50 GB | gp3-max | 3,161±1.0% | 13,979 | 10.13±1.0% | 59.11±0.1% | 61.61±0.1% | 1,580 | 395 |