p2.16xlarge
⚡ 16x GPUPart of the P2 Accelerated Computing family — Hardware accelerators (GPUs, FPGAs) for specialized compute.
Name:
pGPU Deep Learning & HPCp = GPU Deep Learning & HPCHigh-end NVIDIA GPUs (Blackwell, Hopper, Ampere, Volta) built for frontier AI/LLM training, HPC simulations, and generative AI.22nd Generation2 = 2nd GenerationAWS 2nd generation hardware platform with updated CPU/system architecture.16xlarge16 Extra Large16xlarge = 16 Extra Large64 vCPUs • 732.0 GiB RAM • EBS only • 20 Gigabit Network • 16x NVIDIA Tesla K80Price (Linux)
$14.4000/ hr
~$10,512.00/mo
Compute
64vCPUs
64 threads • x86_64
Memory
732.0GiB
11.4 GiB / vCPU
Value
$0.2250/ vCPU-hr
$0.0197/GiB-hr
Network
20 Gigabit
20 Gigabit
Storage & EBS
EBSOnly
EBS-optimized
Technical Specifications & Limits
Hardware specs, container thresholds, and system limitsCompute & Processor
ProcessorIntel Xeon E5-2686 v4 (Broadwell)
Architecturex86_64 (x86_64)
vCPUs / Threads64 vCPUs
Physical Cores—
Clock Speed2.3 GHz
Intel AVX / AVX2AVX & AVX2 Supported
Intel Turbo BoostYes
Bare MetalNo (Virtualized)
Performance & Benchmarks
CoreMark ScoreNot benchmarked
CoreMark / vCPU—
FFmpeg Transcoding—
CoreMark per Dollar—
Cost per vCPU-hr$0.2250
Cost per GiB-hr$0.0197
Containers & Kubernetes
EKS Max Pods—
Max ECS Tasks—
Max ENIs—
IPv4/v6 per ENI—
Total Addressable IPs—
CNI Prefix DelegationStandard CNI
Networking & Bandwidth
Network Performance20 Gigabit
Baseline Bandwidth—
Burst / Peak Bandwidth—
Enhanced NetworkingYes
EFA SupportNo
IPv6 SupportNative Dual-Stack Supported
Storage & EBS Optimization
Storage ConfigurationEBS only
Local NVMe DiskNone (EBS only)
EBS Baseline Throughput—
EBS Max Throughput—
EBS Baseline IOPS—
EBS Max / Burst IOPS—
Dedicated EBS Throughput10000 Mbps
GPU & Hardware Acceleration
GPU ModelNVIDIA Tesla K80
Accelerator Count16x GPU(s)
Total GPU Memory—
Memory ArchitectureGDDR / HBM
CUDA Compute Capabilityv3.7
Acceleration RuntimesCUDA, TensorRT, cuDNN, PyTorch, PyJAX
Platform & Architecture Details
Hypervisor / SystemXen Hypervisor
Nitro EnclavesNot Supported
Normalization Factor128
Generation StatusCurrent Generation (Recommended)
Date Introduced2016-09-29
Regional Pricing Matrix
Live on-demand rates across all global AWS cloud regionsLowest Price Region
$14.4000/ hr
US East (N. Virginia) (us-east-1)Highest Price Region
$27.4880/ hr
Asia Pacific (Mumbai) (ap-south-1)Global Fleet Average
$20.4553/ hr
Across all active regionsRegional Coverage
11regions
Available worldwide| Region Name | Region Code | On-Demand Price |
|---|---|---|
Asia Pacific (Mumbai)Highest | ap-south-1 | $27.4880 |
Asia Pacific (Seoul) | ap-northeast-2 | $23.4400 |
Asia Pacific (Singapore)Highest | ap-southeast-1 | $27.4880 |
Asia Pacific (Sydney) | ap-southeast-2 | $24.6720 |
Asia Pacific (Tokyo) | ap-northeast-1 | $24.6720 |
AWS GovCloud (US-West) | us-gov-west-1 | $17.2800 |
EU (Frankfurt) | eu-central-1 | $21.2160 |
EU (Ireland) | eu-west-1 | $15.5520 |
US East (N. Virginia)Lowest | us-east-1 | $14.4000 |
US East (Ohio)Lowest | us-east-2 | $14.4000 |
US West (Oregon)Lowest | us-west-2 | $14.4000 |
| Region Name | Region Code | On-Demand Price |
|---|---|---|
Asia Pacific (Mumbai)Highest | ap-south-1 | $30.4320 |
Asia Pacific (Seoul) | ap-northeast-2 | $26.3840 |
Asia Pacific (Singapore)Highest | ap-southeast-1 | $30.4320 |
Asia Pacific (Sydney) | ap-southeast-2 | $27.6160 |
Asia Pacific (Tokyo) | ap-northeast-1 | $27.6160 |
AWS GovCloud (US-West) | us-gov-west-1 | $20.2240 |
EU (Frankfurt) | eu-central-1 | $24.1600 |
EU (Ireland) | eu-west-1 | $18.4960 |
US East (N. Virginia)Lowest | us-east-1 | $17.3440 |
US East (Ohio)Lowest | us-east-2 | $17.3440 |
US West (Oregon)Lowest | us-west-2 | $17.3440 |
No regions matched your search filter.