c8in.32xlarge vs trn1n.32xlarge
| c8in.32xlarge | trn1n.32xlarge |
| Compute optimized | Machine Learning ASIC |
| High-performance web servers, scientific modeling, batch processing, distributed analytics, HPC, machine/deep learning inference, ad serving, scalable multiplayer gaming, and video encoding. c – Compute optimized 8 – Generation i – Intel processor n – Network and EBS optimized 32xlarge – Size | Machine Learning (ML) training trn – AWS Trainium 1 – Generation n – Network and EBS optimized 32xlarge – Size |
| 128 | 128 |
| 256 GiB | 512 GiB |
| x86_64 | x86_64 |
| Intel Xeon Scalable (Granite Rapids) | Intel Xeon Scalable (Icelake) |
| 3.9 GHz | 3.5 GHz |
| 2 | 2 |
| no | no |
| 16 | |
| no | no |
| 512 GiB | |
| no | no |
| c8in.12xlarge, c8in.16xlarge, c8in.24xlarge, c8in.2xlarge, c8in.32xlarge, c8in.48xlarge, c8in.4xlarge, c8in.8xlarge, c8in.96xlarge, c8in.large, c8in.metal-48xl, c8in.metal-96xl, c8in.xlarge | trn1n.32xlarge |
| c6in.32xlarge, c8in.32xlarge | |
| c8i.32xlarge, c8ib.32xlarge, c8id.32xlarge | trn1.32xlarge |
| ml.trn1n.32xlarge | |
| no | no |
| no | no |
| no | no |
| EBS only | 4 x 1900 NVMe SSD |
| default | default |
| supported | supported |
| required | required |
| 40000 MBps | 81920 MBps |
| 40000 Mbps | 80000 Mbps |
| 5000 Mbps | 10000 Mbps |
| 160000 IOPS | 260000 IOPS |
| 160000 IOPS | 260000 IOPS |
| 40000 Mbps | 80000 Mbps |
| 5000 Mbps | 10000 Mbps |
| 200 Gigabit | 1600 Gigabit |
| yes | yes |
| 16 | 80 |
| 1 | 16 |
| 64 | 50 |
| 64 | 50 |
| yes | yes |
| required | required |
| no | yes |
| 16 | |
| yes | yes |
| yes | no |
| 1.00x | 2.59x |
Regional Prices
Geography | Region | c8in.32xlarge | trn1n.32xlarge |
|---|---|---|---|
| Europe | Europe (Spain) (eu-south-2) | 9.8304 | n/a |
| Hidden | Hidden (hidden-1) | 10.9670 | n/a |
| US | US East (Ohio) (us-east-2) | n/a | 24.7800 |
| US | US East (Virginia) (us-east-1) | 8.7091 | 24.7800 |
| US | US West (Oregon) (us-west-2) | 8.7091 | 24.7800 |