Star
Optimize Kubernetes clusters with Pod Efficiency Metrics ->
Slack
Star

g3.16xlarge

The g3.16xlarge instance is in the gpu instance family with 64 vCPUs, 488.0 GiB of memory and 25 Gibps of bandwidth starting at $4.56 per hour.

paid

Pricing

On Demand

Spot

1 Yr Reserved

3 Yr Reserved

$1696.52 per month (-49%) with Autopilot

Learn more
dns

Family Sizes

Size vCPUs Memory (GiB)
g3.4xlarge 16 122
g3.8xlarge 32 244
g3.16xlarge 64 488
dns

Instance Variants

Variant
g3s
g3
info

Instance Details

Compute Value
vCPUs 64
Memory (GiB) 488.0
Memory per vCPU (GiB) 7.62
Physical Processor Intel Xeon E5-2686 v4 (Broadwell)
Clock Speed (GHz) 2.3
CPU Architecture x86_64
GPU 4
GPU Architecture NVIDIA Tesla M60
Video Memory (GiB) 32
GPU Compute Capability (?) 5.2
FPGA 0
Networking Value
Network Performance (Gibps) 25
Enhanced Networking True
IPV6 True
Placement Group (?) True
Storage Value
EBS Optimized True
Max Bandwidth (Mbps) on (EBS) 14000
Max Throughput (MB/s) on EBS 1750.0
Max I/O Operations/second (IOPS) 80000
Baseline Bandwidth (Mbps) on (EBS) 14000
Baseline Throughput (MB/s) on EBS 1750.0
Baseline I/O Operations/second (IOPS) 80000
Amazon Value
Generation current
Instance Type g3.16xlarge
Family GPU instance
Name G3 Graphics GPU 16xlarge
Elastic Map Reduce (EMR) True

See a data problem? Open a ticket.



Meet us at re:Invent 2023