h1.2xlarge vs g7.2xlarge
| h1.2xlarge | g7.2xlarge |
| Storage optimized | GPU instance |
| Amazon EMR workloads, distributed file systems (HDFS, MapR-FS), network file systems, log/data processing apps (like Apache Kafka), and big data clusters. h – HDD / hard disk throughput optimized 1 – Generation 2xlarge – Size | Machine Learning (ML), 3D visualizations, graphics-heavy remote workstations, 3D rendering, application streaming, video encoding, and server-side graphics tasks. g – GPU (Graphics Processing Unit) 7 – Generation 2xlarge – Size |
| 8 | 8 |
| 32 GiB | 32 GiB |
| x86_64 | x86_64 |
| Intel Xeon E5-2686 v4 (Broadwell) | Intel Xeon Scalable (Granite Rapids) |
| 2.3 GHz | 3.9 GHz |
| 2 | 1 |
| no | yes |
| RTX PRO 4500 | |
| NVIDIA | |
| 1 | |
| 32 GiB | |
| 32 GiB | |
| no | no |
| 32 GiB | |
| no | no |
| h1.16xlarge, h1.2xlarge, h1.4xlarge, h1.8xlarge | g7.12xlarge, g7.24xlarge, g7.2xlarge, g7.48xlarge, g7.4xlarge, g7.8xlarge |
| g2.2xlarge, g5.2xlarge, g6.2xlarge, g7.2xlarge | |
| g7e.2xlarge | |
| ml.g7.2xlarge | |
| no | no |
| no | no |
| no | no |
| 31 | |
| 1 x 2000 HDD | 1x440 GB NVMe SSD |
| default | default |
| supported | supported |
| unsupported | required |
| 1750 MBps | up to 8000 MBps |
| 1750 Mbps | 4000 Mbps |
| 218.75 Mbps | 500 Mbps |
| 12000 IOPS | 16000 IOPS |
| 12000 IOPS | 32000 IOPS |
| 1750 Mbps | 8000 Mbps |
| 218.75 Mbps | 1000 Mbps |
| Up to 10 Gigabit | Up to 60 Gigabit |
| yes | yes |
| 4 | 4 |
| 1 | 1 |
| 15 | 64 |
| 15 | 64 |
| yes | yes |
| supported | required |
| no | no |
| no | yes |
| no | no |
| 1.00x | 5.31x |
Regional Prices
Geography | Region | h1.2xlarge | g7.2xlarge |
|---|---|---|---|
| Hidden | Hidden (hidden-1) | 0.5190 | 2.6460 |
| US | US East (Ohio) (us-east-2) | 0.4680 | 2.5200 |
| US | US East (Virginia) (us-east-1) | 0.4680 | 2.5200 |
| US | US West (Oregon) (us-west-2) | 0.4680 | 2.5200 |
Related Comparisons
Other instances worth weighing against these.