Amazon Elastic Graphics
Important
Amazon Elastic Graphics will reach end of life on January 8, 2024. Starting September 5, 2023 the service is no longer accepting new customer
accounts. If you are currently using Amazon Elastic Graphics, we recommend that you migrate your workloads to Amazon EC2 G4ad instances, G4dn instances, or G5 instances at your
earliest convenience. If you have concerns or questions regarding end of life of the service, contact Amazon Web Services Support
Amazon Elastic Graphics provides flexible, low-cost, and high performance graphics acceleration for your Windows instances. Elastic Graphics accelerators come in multiple sizes and are a low-cost alternative to using GPU graphics instance types (such as G2 and G3). You have the flexibility to choose an instance type that meets the compute, memory, and storage needs of your application. Then, choose the accelerator for your instance that meets the graphics requirements of your workload.
Elastic Graphics is suited for applications that require a small or intermittent amount of additional graphics acceleration, and that use OpenGL graphics support. If you need access to full, directly attached GPUs and use of DirectX, CUDA, or Open Computing Language (OpenCL) parallel computing frameworks, use an accelerated computing instance type instance instead. For more information, see Windows accelerated computing instances.
Contents
Elastic Graphics basics
To use Elastic Graphics, launch a Windows instance and specify an accelerator type for the instance during launch. Amazon finds available Elastic Graphics capacity and establishes a network connection between your instance and the Elastic Graphics accelerator.
Note
Bare metal instances are not supported.
Elastic Graphics accelerators are available in the following Amazon Regions: us-east-1,
us-east-2, us-west-2, ap-northeast-1,
ap-southeast-1, ap-southeast-2,
eu-central-1, and eu-west-1.
The following instance types support Elastic Graphics accelerators:
-
General purpose: M3, M4, M5, M5d, M5dn, M5n, T2, and T3
Note
Only
t2.mediumand larger andt3.mediumand larger are supported. -
Compute optimized: C3, C4, C5, C5a, C5ad, C5d, and C5n
-
Memory optimized: R3, R4, R5, R5d, R5dn, R5n, X1, X1e, and z1d
-
Storage optimized: D2, D3, D3en, H1, I3, and I3en
-
Accelerated computing: P2, P3, and P3dn
The following Elastic Graphics accelerators are available. You can attach any Elastic Graphics accelerator to any supported instance type.
| Elastic Graphics accelerator | Graphics memory (GB) |
|---|---|
eg1.medium |
1 |
eg1.large |
2 |
eg1.xlarge |
4 |
eg1.2xlarge |
8 |
An Elastic Graphics accelerator does not form part of the hardware of your instance. Instead, it is network-attached through a network interface, known as the Elastic Graphics network interface. When you launch or restart an instance with graphics acceleration, the Elastic Graphics network interface is created in your VPC for you.
The Elastic Graphics network interface is created in the same subnet and VPC as your instance and is assigned a private IPv4 address from that subnet. The accelerator attached to your Amazon EC2 instance is allocated from a pool of available accelerators in the same Availability Zone as your instance.
Elastic Graphics accelerators support the API standards for OpenGL 4.3 API and earlier, which can be used for batch applications or 3D-graphics acceleration. An Amazon-optimized OpenGL library on your instance detects the attached accelerator. It directs OpenGL API calls from your instance to the accelerator, which then processes the requests and returns the results. Traffic between the instance and the accelerator uses the same bandwidth as the instance's network traffic so we recommend that you have adequate network bandwidth available. Consult your software vendor for any OpenGL compliance and version questions.
By default, the default security group for your VPC is associated with the Elastic Graphics network interface. The Elastic Graphics network traffic uses the TCP protocol and port 2007. Ensure that the security group for your instance allows for this. For more information, see Configure your security groups.
Pricing for Elastic Graphics
You are charged for each second that an Elastic Graphics accelerator is attached to an
instance in the running state when the accelerator is in the
Ok state. You are not charged for an accelerator attached to an
instance that is in the pending, stopping,
stopped, shutting-down, or terminated
state. You are also not charged when an accelerator is in the Unknown
or Impaired state.
Pricing for accelerators is available at On-Demand rates only. You can attach an accelerator to a Reserved Instance or Spot Instance, however, the On-Demand price for the accelerator applies.
For more information, see Amazon Elastic Graphics Pricing
Elastic Graphics limitations
Before you start using Elastic Graphics accelerators, be aware of the following limitations:
-
You can attach accelerators only to Windows instances with Microsoft Windows Server 2012 R2 or later. Linux instances are currently not supported.
-
You can attach one accelerator to an instance at a time.
-
You can attach an accelerator only during instance launch. You cannot attach an accelerator to an existing instance.
-
You can't hibernate an instance with an attached accelerator.
-
You can't share an accelerator between instances.
-
You can't detach an accelerator from an instance or transfer it to another instance. If you no longer require an accelerator, you must terminate your instance. To change the accelerator type, create an AMI from your instance, terminate the instance, and launch a new instance with a different accelerator specification.
-
The only supported versions of the OpenGL API are 4.3 and earlier. DirectX, CUDA, and OpenCL are not supported.
-
The Elastic Graphics accelerator is not visible or accessible through the device manager of your instance.
-
You can't reserve or schedule accelerator capacity.