inf2.8xlarge vs p3.8xlarge
| inf2.8xlarge | p3.8xlarge |
| Machine Learning ASIC | GPU instance |
| Machine Learning (ML) inference inf – AWS Inferentia 2 – Generation 8xlarge – Size | GPU intensive tasks like machine/deep learning, high-performance computing p – Performance / GPU (Graphics Processing Unit) accelerated 3 – Generation 8xlarge – Size |
| 32 | 32 |
| 128 GiB | 244 GiB |
| x86_64 | n/a |
| AMD EPYC 7R13 Processor | Intel Xeon E5-2686 v4 (Broadwell) |
| 3.6 GHz | |
| 2 | 0 |
| no | no |
| 1 | 4 |
| yes | no |
| Inferentia2 | |
| AWS | |
| 1 | |
| 32 GiB | 64 GiB |
| no | no |
| inf2.24xlarge, inf2.48xlarge, inf2.8xlarge, inf2.xlarge | p3.16xlarge, p3.2xlarge, p3.8xlarge |
| p2.8xlarge, p3.8xlarge | |
| ml.inf2.8xlarge | |
| no | no |
| no | no |
| no | no |
| 97 | |
| EBS only | EBS only |
| default | |
| supported | |
| required | |
| 10240 MBps | 7000 MBps |
| 10000 Mbps | |
| 1250 Mbps | |
| 40000 IOPS | |
| 40000 IOPS | |
| 10000 Mbps | |
| 1250 Mbps | |
| Up to 25 Gigabit | 10 Gigabit |
| yes | yes |
| 8 | |
| 1 | |
| 30 | |
| 30 | |
| yes | no |
| required | |
| no | no |
| yes | no |
| no | no |
| 1.00x | 5.69x |
Regional Prices
Geography | Region | inf2.8xlarge | p3.8xlarge |
|---|---|---|---|
| Hidden | Hidden (hidden-1) | 2.1647 | 13.2200 |
| Hidden | Hidden (hidden-10) | 3.3454 | n/a |
| Hidden | Hidden (hidden-2) | 2.4598 | 13.4640 |
| Hidden | Hidden (hidden-3) | 2.5582 | 14.3560 |
| Hidden | Hidden (hidden-4) | 2.5582 | 14.6880 |
| Hidden | Hidden (hidden-5) | 2.7550 | 15.2920 |
| Hidden | Hidden (hidden-6) | 2.7550 | 16.7760 |
| Hidden | Hidden (hidden-7) | 2.9518 | 16.9360 |
| Hidden | Hidden (hidden-8) | 2.9518 | 16.9360 |
| Hidden | Hidden (hidden-9) | 2.9518 | 16.9360 |
| US | US East (Ohio) (us-east-2) | 1.9679 | 12.2400 |
| US | US East (Virginia) (us-east-1) | 1.9679 | 12.2400 |
| US | US West (Oregon) (us-west-2) | 1.9679 | 12.2400 |
Activate subscription to see all regions and unlock all features
Related Comparisons
Other instances worth weighing against these.