g7.2xlarge vs g7e.2xlarge

Accelerators
Instance Type	g7.2xlarge	g7e.2xlarge
Instance Family	GPU instance	GPU instance
Details	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	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 e – Extra storage or memory 2xlarge – Size
vCPUs	8	8
Memory	32 GiB	64 GiB
CPU Architecture	x86_64	x86_64
Processor	Intel Xeon Scalable (Granite Rapids)	Intel Xeon Scalable (Emerald Rapids)
Clock Speed	3.9 GHz	2.4 GHz
Threads Per Core	1	2
GPU	yes	yes
GPU Name	RTX PRO 4500	RTX PRO Server 6000
GPU Manufacturer	NVIDIA	NVIDIA
GPUs	1	1
GPU Memory	32 GiB	96 GiB
GPU Total Memory	32 GiB	96 GiB
Inference Accelerator	no	no
Accelerator Memory	32 GiB	96 GiB
FPGA	no	no
Instances
Sizes	g7.12xlarge, g7.24xlarge, g7.2xlarge, g7.48xlarge, g7.4xlarge, g7.8xlarge	g7e.12xlarge, g7e.24xlarge, g7e.2xlarge, g7e.48xlarge, g7e.4xlarge, g7e.8xlarge
Generations	g2.2xlarge, g5.2xlarge, g6.2xlarge, g7.2xlarge	g6e.2xlarge, g7e.2xlarge
Variants	g7e.2xlarge	g7.2xlarge
SageMaker	ml.g7.2xlarge	ml.g7e.2xlarge
Free Tier	no	no
Burstable	no	no
Hibernation	no	no
Instance Storage	1x600 GB NVMe SSD	1 x 1900 NVMe SSD
EBS (Elastic Block Store)
EBS Optimized	default	default
Encryption	supported	supported
NVMe	required	required
Dedicated Throughput	up to 8000 MBps	up to 5000 MBps
Optimized Baseline Bandwidth	4000 Mbps	2000 Mbps
Optimized Baseline Throughput	500 Mbps	250 Mbps
Optimized Baseline Performance	16000 IOPS	8000 IOPS
Optimized Max Performance	32000 IOPS	20000 IOPS
Optimized Max Bandwidth	8000 Mbps	5000 Mbps
Optimized Max Throughput	1000 Mbps	625 Mbps
Networking
Networking Performance	60 Gigabit	50 Gigabit
Enhanced Networking	yes	yes
Max Interfaces	4	4
Max Cards	1	1
IPv4 Addresses per Interface	64	64
IPv6 Addresses per Interface	64	64
IPv6 Support	yes	yes
ENA Support	required	required
EFA Support	no	no
Encryption in Transit Support	yes	yes
ENA SRD Support	no	no
Avg Price Diff	1.00x	1.77x

Regional Prices

US Dollar ($)

Per Hour

Linux

Shared Tenancy

On Demand

Geography	Region	g7.2xlarge	g7e.2xlarge
Hidden	Hidden (hidden-1)	n/a	3.5438
Hidden	Hidden (hidden-2)	n/a	4.1348
Hidden	Hidden (hidden-3)	n/a	4.8516
Hidden	Hidden (hidden-4)	n/a	4.8775
Hidden	Hidden (hidden-5)	n/a	5.4930
Hidden	Hidden (hidden-6)	n/a	5.7192
Hidden	Hidden (hidden-7)	n/a	5.8453
US	US East (Ohio) (us-east-2)	2.5200	3.3631
US	US East (Virginia) (us-east-1)	2.5200	3.3631
US	US West (Oregon) (us-west-2)	2.5200	3.3631