p2.16xlarge

Amazon EC2 · P2 family

The p2.16xlarge is a gpu & accelerated Amazon EC2 instance type with 64 vCPUs and 732 GiB of memory, 16 × NVIDIA K80, running on Intel Xeon E5-2686 v4 (Broadwell) with 20 Gigabit networking, starting at $14.40 per hour on demand.

gpu & acceleratedx86-64intelgpu
vCPUs
64

dedicated

memory
732 GiB

11.44 GiB per vCPU

gpu
16× K80

192 GiB total

on-demand
$14.40

$10,512/mo · us-east-1

spot
$5.787

−60% vs on-demand

regions
11

across 4 areas

p2.16xlarge price by region

11 regions · hourly list prices in USD · On-demand · Spot · 1-yr Reserved · 3-yr Reserved

US East (N. Virginia)us-east-1cheapestreferenceNorth America$14.40$5.787
US East (Ohio)us-east-2cheapestNorth America$14.40$2.449
US West (Oregon)us-west-2cheapestNorth America$14.40$2.604
EU (Ireland)eu-west-1Europe$15.55$7.344
AWS GovCloud (US)us-gov-west-1North America$17.28–
EU (Frankfurt)eu-central-1Europe$21.22–
Asia Pacific (Seoul)ap-northeast-2Asia$23.44–
Asia Pacific (Sydney)ap-southeast-2Oceania$24.67$9.172
Asia Pacific (Tokyo)ap-northeast-1Asia$24.67$5.687
Asia Pacific (Mumbai)ap-south-1Asia$27.49$20.75
Asia Pacific (Singapore)ap-southeast-1Asia$27.49$13.44

Specifications

compute

vCPUs
64
Memory
732 GiB
Memory per vCPU
11.44 GiB
Processor
Intel Xeon E5-2686 v4 (Broadwell)
CPU vendor
Intel
Architecture
x86-64
Clock speed
2.3 GHz
Burstable CPU
No

storage & network

Local storage
Network storage only
Storage detail
EBS only
Network
20 Gigabit
Peak bandwidth
20 Gbps

accelerators

Count
16
Memory (total)
192 GiB
Architecture
Kepler
Vendor
NVIDIA

aws details

API name
p2.16xlarge
Family
P2
Category
GPU & accelerated
Generation
Current
Regions
11 (Asia, Europe, North America, Oceania)
Cheapest region
US East (N. Virginia) · $14.40/hr
Dedicated EBS throughput
10000 Mbps
ECU
201
Enhanced networking
Yes
Normalization size factor
128
Processor features
Intel AVX; Intel AVX2; Intel Turbo

P2 family sizes

all about P2→
SizevCPUMemoryAcceleratorLocal storageOn-demandMonthlySpot
p2.xlarge461 GiB1 × NVIDIA K80–$0.90$657$0.4249
p2.8xlarge32488 GiB8 × NVIDIA K80–$7.20$5,256$2.999
p2.16xlarge64732 GiB16 × NVIDIA K80–$14.40$10,512$5.787

p2.16xlarge FAQ

How much does p2.16xlarge cost?

p2.16xlarge costs $14.40 per hour on demand with Linux in US East (N. Virginia) (us-east-1), about $10,512 per month. Across 11 regions the Linux on-demand price ranges from $14.40 in US East (N. Virginia) to $27.49 in Asia Pacific (Singapore). Spot capacity costs $5.787 per hour (60% less), but can be reclaimed. With Windows it costs $17.34 per hour.

How many vCPUs and how much memory does p2.16xlarge have?

p2.16xlarge has 64 vCPUs and 732 GiB of memory (11.44 GiB per vCPU), running on Intel Xeon E5-2686 v4 (Broadwell). It is a gpu & accelerated instance type in the P2 family.

What GPU does p2.16xlarge have?

p2.16xlarge has 16 × NVIDIA K80 with 192 GiB of accelerator memory in total.