a1.4xlarge vs p5.4xlarge
| a1.4xlarge | p5.4xlarge |
| General purpose | GPU instance |
| Obsolete. Not recommended for use Web servers, containerized microservices, caching fleets, distributed data stores, and dev environments. a – ARM AWS Graviton Processor 1 – Generation 4xlarge – Size | GPU intensive tasks like machine/deep learning, high-performance computing p – Performance / GPU (Graphics Processing Unit) accelerated 5 – Generation 4xlarge – Size |
| 16 | 16 |
| 32 GiB | 256 GiB |
| arm64 | x86_64 |
| AWS Graviton Processor | AMD EPYC 7R13 Processor |
| 2.3 GHz | 3.6 GHz |
| 1 | 2 |
| no | yes |
| H100 | |
| NVIDIA | |
| 1 | |
| 80 GiB | |
| 80 GiB | |
| no | no |
| 803 GiB | |
| no | no |
| a1.2xlarge, a1.4xlarge, a1.large, a1.medium, a1.metal, a1.xlarge | p5.48xlarge, p5.4xlarge |
| ml.p5.4xlarge | |
| no | no |
| no | no |
| no | no |
| EBS only | 1 x 3840 GB SSD |
| default | default |
| supported | supported |
| required | required |
| 3500 MBps | 10240 MBps |
| 3500 Mbps | 10000 Mbps |
| 437.5 Mbps | 1250 Mbps |
| 20000 IOPS | 32500 IOPS |
| 20000 IOPS | 32500 IOPS |
| 3500 Mbps | 10000 Mbps |
| 437.5 Mbps | 1250 Mbps |
| Up to 10 Gigabit | 100 Gigabit |
| yes | yes |
| 8 | 4 |
| 1 | 1 |
| 30 | 30 |
| 30 | 30 |
| yes | yes |
| required | required |
| no | yes |
| 1 | |
| no | yes |
| no | yes |
| 1.00x | 19.58x |
Regional Prices
Geography | Region | a1.4xlarge | p5.4xlarge |
|---|---|---|---|
| Asia Pacific | Asia Pacific (Jakarta) (ap-southeast-3) | n/a | n/a |
| Asia Pacific | Asia Pacific (Mumbai) (ap-south-1) | 0.4080 | n/a |
| Hidden | Hidden (hidden-1) | 0.4080 | n/a |
| Hidden | Hidden (hidden-2) | 0.4608 | 6.8800 |
| Hidden | Hidden (hidden-3) | 0.4656 | 8.2560 |
| Hidden | Hidden (hidden-4) | 0.4704 | 8.6000 |
| Hidden | Hidden (hidden-5) | 0.5136 | 8.9440 |
| Hidden | Hidden (hidden-6) | 0.5328 | 8.9440 |
| Hidden | Hidden (hidden-7) | n/a | 11.5584 |
| US | US East (Ohio) (us-east-2) | 0.4080 | n/a |
| US | US East (Virginia) (us-east-1) | 0.4080 | n/a |