The p2.16xlarge instance is in the gpu instance family with 64 vCPUs, 732.0 GiB of memory starting at $14.4 per hour.



On Demand


1 Yr Reserved

3 Yr Reserved


Family Sizes

Size vCPUs Memory (GiB)
p2.xlarge 4 61
p2.8xlarge 32 488

Instance Details

Compute Value
vCPUs 64
Memory (GiB) 732.0
Memory per vCPU (GiB) 11.44
Physical Processor Intel Xeon E5-2686 v4 (Broadwell)
Clock Speed (GHz) 2
CPU Architecture x86_64
GPU 16
GPU Architecture NVIDIA Tesla K80
Video Memory (GiB) 192
GPU Compute Capability (?) 3.7
Concerned about cloud costs? Connect your AWS account in under 5 minutes to see savings.
Networking Value
Network Performance (Gibps) 25 Gigabit
Enhanced Networking True
IPV6 True
Placement Group (?) True
Storage Value
EBS Optimized True
Max Bandwidth (Mbps) on (EBS) 10000
Max Throughput (MB/s) on EBS 1250.0
Max I/O Operations/second (IOPS) 65000
Baseline Bandwidth (Mbps) on (EBS) 0
Baseline Throughput (MB/s) on EBS 0
Baseline I/O Operations/second (IOPS) 0
Amazon Value
Generation current
Instance Type p2.16xlarge
Family GPU instance
Name P2 General Purpose GPU 16xlarge
Elastic Map Reduce (EMR) True