Windows accelerated computing instances
Accelerated computing instances use hardware accelerators, or co-processors, to perform some functions, such as floating point number calculations, graphics processing, or data pattern matching, more efficiently than is possible in software running on CPUs. These instances enable more parallelism for higher throughput on compute-intensive workloads.
If you require high processing capability, you'll benefit from using accelerated computing instances, which provide access to hardware-based compute accelerators such as Graphics Processing Units (GPUs).
Contents
GPU instances
GPU-based instances provide access to NVIDIA GPUs with thousands of compute cores. You can use these instances to accelerate scientific, engineering, and rendering applications by leveraging the CUDA or Open Computing Language (OpenCL) parallel computing frameworks. You can also use them for graphics applications, including game streaming, 3-D application streaming, and other graphics workloads.
If your application needs a small amount of additional graphics acceleration, but is better suited for an instance type with different compute, memory, or storage specifications, use an Elastic Graphics accelerator instead. For more information, see Amazon Elastic Graphics.
G4ad and G4dn instances
G4ad instances use AMD Radeon Pro V520 GPUs and 2nd generation AMD EPYC processors, and are well-suited for graphics applications such as remote graphics workstations, game streaming, and rendering that leverage industry-standard APIs such as OpenGL, DirectX, and Vulkan. They provide up to 4 AMD Radeon Pro V520 GPUs, 64 vCPUs, 25 Gbps networking, and 2.4 TB local NVMe-based SSD storage.
G4dn instances use NVIDIA Tesla GPUs and provide a cost-effective, high-performance platform for general purpose GPU computing using the CUDA or machine learning frameworks along with graphics applications using DirectX or OpenGL. These instances provide high- bandwidth networking, powerful half and single-precision floating-point capabilities, along with INT8 and INT4 precisions. Each GPU has 16 GiB of GDDR6 memory, making G4dn instances well-suited for machine learning inference, video transcoding, and graphics applications like remote graphics workstations and game streaming in the cloud.
For more information, see Amazon EC2 G4 Instances
G4dn instances support NVIDIA GRID Virtual Workstation. For more information, see
NVIDIA Marketplace offerings
G3 instances
These instances use NVIDIA Tesla M60 GPUs and provide a cost-effective, high-performance platform for graphics applications using DirectX or OpenGL. G3 instances also provide NVIDIA GRID Virtual Workstation features, such as support for four monitors with resolutions up to 4096x2160, and NVIDIA GRID Virtual Applications. G3 instances are well-suited for applications such as 3D visualizations, graphics-intensive remote workstations, 3D rendering, video encoding, virtual reality, and other server-side graphics workloads requiring massively parallel processing power.
For more information, see Amazon EC2 G3 Instances
G3 instances support NVIDIA GRID Virtual Workstation and NVIDIA GRID Virtual Applications. To activate either of these features, see Activate NVIDIA GRID Virtual Applications.
G2 instances
These instances use NVIDIA GRID K520 GPUs and provide a cost-effective, high-performance platform for graphics applications using DirectX or OpenGL. NVIDIA GRID GPUs also support NVIDIA’s fast capture and encode API operations. Example applications include video creation services, 3D visualizations, streaming graphics-intensive applications, and other server-side graphics workloads.
P3 instances
These instances use NVIDIA Tesla V100 GPUs and are designed for general purpose GPU computing using the CUDA or OpenCL programming models or through a machine learning framework. P3 instances provide high-bandwidth networking, powerful half, single, and double-precision floating-point capabilities, and up to 32 GiB of memory per GPU, which makes them ideal for deep learning, computational fluid dynamics, computational finance, seismic analysis, molecular modeling, genomics, rendering, and other server-side GPU compute workloads. Tesla V100 GPUs do not support graphics mode.
For more information, see Amazon EC2 P3 Instances
P3 instances support NVIDIA NVLink peer to peer transfers. For more information, see
NVIDIA NVLink
P2 instances
P2 instances use NVIDIA Tesla K80 GPUs and are designed for general purpose GPU computing using the CUDA or OpenCL programming models. P2 instances provide high-bandwidth networking, powerful single and double precision floating-point capabilities, and 12 GiB of memory per GPU, which makes them ideal for deep learning, graph databases, high-performance databases, computational fluid dynamics, computational finance, seismic analysis, molecular modeling, genomics, rendering, and other server-side GPU compute workloads.
P2 instances support NVIDIA GPUDirect peer to peer transfers. For more information,
see NVIDIA GPUDirect
Hardware specifications
The following is a summary of the hardware specifications for accelerated computing instances.
| Instance type | Default vCPUs | Memory (GiB) | Accelerators |
|---|---|---|---|
p2.xlarge |
4 | 61 | 1 |
p2.8xlarge |
32 | 488 | 8 |
p2.16xlarge |
64 | 732 | 16 |
p3.2xlarge |
8 | 61 | 1 |
p3.8xlarge |
32 | 244 | 4 |
p3.16xlarge |
64 | 488 | 8 |
p3dn.24xlarge |
96 | 768 | 8 |
g2.2xlarge |
8 | 15 | 1 |
g2.8xlarge |
32 | 60 | 4 |
g3s.xlarge |
4 | 30.5 | 1 |
g3.4xlarge |
16 | 122 | 1 |
g3.8xlarge |
32 | 244 | 2 |
g3.16xlarge |
64 | 488 | 4 |
g4ad.xlarge |
4 | 16 | 1 |
g4ad.2xlarge |
8 | 32 | 1 |
g4ad.4xlarge |
16 | 64 | 1 |
g4ad.8xlarge |
32 | 128 | 2 |
g4ad.16xlarge |
64 | 256 | 4 |
g4dn.xlarge |
4 | 16 | 1 |
g4dn.2xlarge |
8 | 32 | 1 |
g4dn.4xlarge |
16 | 64 | 1 |
g4dn.8xlarge |
32 | 128 | 1 |
g4dn.12xlarge |
48 | 192 | 4 |
g4dn.16xlarge |
64 | 256 | 1 |
g4dn.metal |
96 | 384 | 8 |
f1.2xlarge |
8 | 122 | 1 |
f1.4xlarge |
16 | 244 | 2 |
f1.16xlarge |
64 | 976 | 8 |
For more information about the hardware specifications for each Amazon EC2 instance
type,
see Amazon EC2 Instance Types
For more information about specifying CPU options, see Optimize CPU options.
Instance performance
EBS-optimized instances enable you to get consistently high performance for your EBS volumes by eliminating contention between Amazon EBS I/O and other network traffic from your instance. Some accelerated computing instances are EBS-optimized by default at no additional cost. For more information, see Amazon EBS–optimized instances.
Network performance
You can enable enhanced networking on supported instance types to provide lower latencies, lower network jitter, and higher packet-per-second (PPS) performance. Most applications do not consistently need a high level of network performance, but can benefit from access to increased bandwidth when they send or receive data. For more information, see Enhanced networking on Windows.
The following is a summary of network performance for accelerated computing instances that support enhanced networking.
| Instance type | Network performance | Enhanced networking |
|---|---|---|
f1.4xlarge and smaller | g3.4xlarge |
g3s.xlarge | g4ad.4xlarge and smaller |
p3.2xlarge
|
Up to 10 Gbps † | ENA |
g3.8xlarge | p2.8xlarge |
p3.8xlarge
|
10 Gbps | ENA |
g4ad.8xlarge |
15 Gbps | ENA |
g4dn.4xlarge and smaller
|
Up to 25 Gbps † | ENA |
f1.16xlarge | g3.16xlarge |
g4ad.16xlarge |
p2.16xlarge | p3.16xlarge
|
25 Gbps | ENA |
g4dn.8xlarge | g4dn.12xlarge | g4dn.16xlarge
|
50 Gbps | ENA |
g4dn.metal | p3dn.24xlarge
|
100 Gbps | ENA |
† These instances have a baseline bandwidth and can use a network I/O credit mechanism to burst beyond their baseline bandwidth on a best effort basis. For more information, see instance network bandwidth.
| Instance type | Baseline bandwidth (Gbps) | Burst bandwidth (Gbps) |
|---|---|---|
g4ad.xlarge |
2 | 10 |
g4ad.2xlarge |
4.167 | 10 |
g4ad.4xlarge |
8.333 | 10 |
g4dn.xlarge |
5 | 25 |
g4dn.2xlarge |
10 | 25 |
g4dn.4xlarge |
20 | 25 |
Instance features
The following is a summary of features for accelerated computing instances.
| EBS only | NVMe EBS | Instance store | Placement group | |
|---|---|---|---|---|
|
F1 |
No |
No |
NVMe * |
Yes |
|
G2 |
No |
No |
SSD |
Yes |
|
G3 |
Yes |
No |
No |
Yes |
|
G4ad |
No |
Yes |
NVMe * |
Yes |
|
G4dn |
No |
Yes |
NVMe * |
Yes |
|
P2 |
Yes |
No |
No |
Yes |
|
P3 |
24xlarge: No
All other sizes: Yes |
24xlarge: Yes
All other sizes: No |
24xlarge: NVMe *
|
Yes |
* The root device volume must be an Amazon EBS volume.
For more information, see the following:
Release notes
-
You must launch the instance using an HVM AMI.
-
Instances built on the Nitro System have the following requirements:
-
NVMe drivers must be installed
-
Elastic Network Adapter (ENA) drivers must be installed
The current AWS Windows AMIs meet these requirements.
-
-
GPU-based instances can't access the GPU unless the NVIDIA drivers are installed. For more information, see Install NVIDIA drivers on Windows instances.
-
Launching a bare metal instance boots the underlying server, which includes verifying all hardware and firmware components. This means that it can take 20 minutes from the time the instance enters the running state until it becomes available over the network.
-
To attach or detach EBS volumes or secondary network interfaces from a bare metal instance requires PCIe native hotplug support.
-
Bare metal instances use a PCI-based serial device rather than an I/O port-based serial device. The upstream Linux kernel and the latest Amazon Linux AMIs support this device. Bare metal instances also provide an ACPI SPCR table to enable the system to automatically use the PCI-based serial device. The latest Windows AMIs automatically use the PCI-based serial device.
-
There is a limit of 100 AFIs per Region.
-
There is a limit on the total number of instances that you can launch in a Region, and there are additional limits on some instance types. For more information, see How many instances can I run in Amazon EC2?
in the Amazon EC2 FAQ. -
If you launch a multi-GPU instance with a Windows AMI that was created on a single-GPU instance, Windows does not automatically install the NVIDIA driver for all GPUs. You must authorize the driver installation for the new GPU hardware. You can correct this manually in the Device Manager by opening the Other device category (the inactive GPUs do not appear under Display Adapters). For each inactive GPU, open the context (right-click) menu, choose Update Driver Software, and then choose the default Automatic Update option.
-
When using Microsoft Remote Desktop Protocol (RDP), GPUs that use the WDDM driver model are replaced with a non-accelerated Remote Desktop display driver. We recommend that you use a different remote access tool to access your GPU, such as Teradici Cloud Access Software
, NICE Desktop Cloud Visualization (DCV), or VNC. You can also use one of the GPU AMIs from the AWS Marketplace because they provide remote access tools that support 3D acceleration.
