Updated at: 2022-04-11 GMT+08:00

GPU-accelerated ECSs

GPU-accelerated ECSs provide outstanding floating-point computing capabilities. They are suitable for applications that require real-time, highly concurrent massive computing.

GPU-accelerated ECS Types

Recommended:

Inference-accelerated PI2

Available now: All GPU models except the recommended ones.

Table 1 GPU-accelerated ECSs

Classification

ECS Type

GPU

CUDA Cores per GPU

Single-GPU Performance

Application Scenario

Remarks

Graphics-accelerated

G5

NVIDIA V100

5120

  • 14 TFLOPS of single-precision floating-point computing
  • 7 TFLOPS of double-precision floating-point computing
  • 112 TFLOPS Tensor Cores for deep learning acceleration

Cloud desktop, image rendering, 3D visualization, and heavy-load graphics design

Remote login on the management console is unavailable. To log in to such an ECS, use VNC or third-party VDI.

Computing-accelerated

P2v

NVIDIA V100 NVLink (GPU passthrough)

5120

  • 15.7 TFLOPS of single-precision floating-point computing
  • 7.8 TFLOPS of double-precision floating-point computing
  • 125 TFLOPS Tensor Cores for deep learning acceleration
  • 300 GiB/s NVLINK

Machine learning, deep learning, inference training, scientific computing, seismic analysis, computing finance, rendering, multimedia encoding and decoding

None

Inference-accelerated

PI2

NVIDIA T4 (GPU passthrough)

2560

  • 8.1 TFLOPS of single-precision floating-point computing
  • 130 INT8 TOPS
  • 260 INT4 TOPS

Machine learning, deep learning, inference training, scientific computing, seismic analysis, computing finance, rendering, multimedia encoding and decoding

None

Inference-accelerated

PI1

NVIDIA P4 (GPU passthrough)

2560

5.5 TFLOPS of single-precision floating-point computing

Machine learning, deep learning, inference training, scientific computing, seismic analysis, computing finance, rendering, multimedia encoding and decoding

None

Images Supported by GPU-accelerated ECSs

Table 2 Image list

Classification

ECS Type

Supported Image

Graphics-accelerated

G5

Windows Server 2016 Standard 64bit

Windows Server 2012 R2 Standard 64bit

CentOS 7.5 64bit

Computing-accelerated

P2v

Windows Server 2016 Standard 64bit

Windows Server 2012 R2 Standard 64bit

Ubuntu 16.04 64bit

CentOS 7.4 64bit

EulerOS 2.2 64bit

Inference-accelerated

PI2

Windows Server 2016 Standard 64bit

Ubuntu 16.04 64bit

CentOS 7.5 64bit

Inference-accelerated

PI1

Ubuntu 16.04 64bit

Ubuntu 14.04 64bit

CentOS 7.3 64bit

Graphics-accelerated Enhancement G5

Overview

G5 ECSs use NVIDIA Tesla V100 GPUs and support DirectX, OpenGL, and Vulkan. These ECSs provide 16 GiB of GPU memory and up to 4096 x 2160 resolution, meeting requirements on professional graphics processing.

Select your desired GPU-accelerated ECS type and specifications.

Specifications

Table 3 G5 ECS specifications

Flavor

vCPUs

Memory (GiB)

Maximum/Assured Bandwidth (Gbit/s)

Max. PPS (10,000)

NIC Multi-Queue

GPUs

GPU Memory (GiB)

Virtualization Type

g5.8xlarge.4

32

128

25/15

200

16

1 x V100

16

KVM

A g5.8xlarge.4 ECS exclusively uses a V100 GPU for professional graphics acceleration. Such an ECS can be used for heavy-load CPU inference.

G5 ECS Features
  • CPU: 2nd Generation Intel® Xeon® Scalable 6278 processors (2.6 GHz of base frequency and 3.5 GHz of turbo frequency), or Intel® Xeon® Scalable 6151 processors (3.0 GHz of base frequency and 3.4 GHz of turbo frequency)
  • Graphics acceleration APIs
    • DirectX 12, Direct2D, DirectX Video Acceleration (DXVA)
    • OpenGL 4.5
    • Vulkan 1.0
  • CUDA and OpenCL
  • NVIDIA V100 GPUs
  • Graphics acceleration applications
  • Heavy-load CPU inference
  • Application flow identical to common ECSs
  • Automatic scheduling of G5 ECSs to AZs where NVIDIA V100 GPUs are used
  • A maximum specification of 16 GiB of GPU memory and 4096 x 2160 resolution for processing graphics and videos

Supported Common Software

G5 ECSs are used in graphics acceleration scenarios, such as video rendering, cloud desktop, and 3D visualization. If the software relies on GPU DirectX and OpenGL hardware acceleration, use G5 ECSs. G5 ECSs support the following commonly used graphics processing software:
  • AutoCAD
  • 3DS MAX
  • MAYA
  • Agisoft PhotoScan
  • ContextCapture
  • Smart3D 3D modeling software

Notes

  • G5 ECSs support the following OSs:
    • Windows Server 2016 Standard 64bit
    • Windows Server 2012 R2 Standard 64bit
    • CentOS 7.5 64bit
  • A G5 ECS requires the configuration of a GRID license after the ECS is created.
  • G5 ECSs created using a public image have had the GRID driver of a specific version installed by default. However, you need to purchase and configure the GRID license by yourself. Ensure that the GRID driver version meets service requirements.

    For instructions about how to configure a GRID license, see Installing a GRID Driver on a GPU-accelerated ECS.

  • If a G5 ECS is created using a private image, make sure that the GRID driver was installed during the private image creation. If not, install the driver for graphics acceleration after the ECS is created.

    For details, see Installing a GRID Driver on a GPU-accelerated ECS.

Computing-accelerated P2s

Overview

P2s ECSs use NVIDIA Tesla V100 GPUs to provide flexibility, high-performance computing, and cost-effectiveness. P2s ECSs provide outstanding general computing capabilities and have strengths in AI-based deep learning, scientific computing, Computational Fluid Dynamics (CFD), computing finance, seismic analysis, molecular modeling, and genomics.

Specifications

Table 4 P2s ECS specifications

Flavor

vCPUs

Memory (GiB)

Maximum/Assured Bandwidth (Gbit/s)

Max. PPS (10,000)

NIC Multi-Queue

Maximum NICs

GPUs

GPU Connection

GPU Memory (GiB)

Virtualization Type

p2s.2xlarge.8

8

64

10/4

50

4

4

1 x V100

PCIe Gen3

1 x 32 GiB

KVM

p2s.4xlarge.8

16

128

15/8

100

8

8

2 x V100

PCIe Gen3

2 x 32 GiB

KVM

p2s.8xlarge.8

32

256

25/15

200

16

8

4 x V100

PCIe Gen3

4 x 32 GiB

KVM

p2s.16xlarge.8

64

512

30/30

400

32

8

8 x V100

PCIe Gen3

8 x 32 GiB

KVM

P2s ECS Features
  • CPU: 2nd Generation Intel® Xeon® Scalable 6278 processors (2.6 GHz of base frequency and 3.5 GHz of turbo frequency), or Intel® Xeon® Scalable 6151 processors (3.0 GHz of base frequency and 3.4 GHz of turbo frequency)
  • Up to eight NVIDIA Tesla V100 GPUs on an ECS
  • NVIDIA CUDA parallel computing and common deep learning frameworks, such as TensorFlow, Caffe, PyTorch, and MXNet
  • 14 TFLOPS of single-precision computing and 7 TFLOPS of double-precision computing
  • NVIDIA Tensor cores with 112 TFLOPS of single- and double-precision computing for deep learning
  • Up to 30 Gbit/s of network bandwidth on a single ECS
  • 32 GiB of HBM2 GPU memory with a bandwidth of 900 Gbit/s
  • Comprehensive basic capabilities

    Networks are user-defined, subnets can be divided, and network access policies can be configured as needed. Mass storage is used, elastic capacity expansion as well as backup and restoration are supported to make data more secure. Auto Scaling allows you to add or reduce the number of ECSs quickly.

  • Flexibility

    Similar to other types of ECSs, P2s ECSs can be provisioned in a few minutes.

  • Excellent supercomputing ecosystem

    The supercomputing ecosystem allows you to build up a flexible, high-performance, cost-effective computing platform. A large number of HPC applications and deep-learning frameworks can run on P2s ECSs.

Supported Common Software

P2s ECSs are used in computing acceleration scenarios, such as deep learning training, inference, scientific computing, molecular modeling, and seismic analysis. If the software is required to support GPU CUDA, use P2s ECSs. P2s ECSs support the following commonly used software:
  • Common deep learning frameworks, such as TensorFlow, Caffe, PyTorch, and MXNet
  • CUDA GPU rendering supported by RedShift for Autodesk 3dsMax and V-Ray for 3ds Max
  • Agisoft PhotoScan
  • MapD
Notes
  • P2s ECSs support the following OSs:
    • Windows Server 2016 Standard 64bit
    • Windows Server 2012 R2 Standard 64bit
    • Ubuntu Server 16.04 64bit
    • CentOS 7.7 64bit
    • CentOS 7.4 64bit
  • By default, P2s ECSs created using a public image have the Tesla driver installed.
  • If a P2s ECS is created using a private image, make sure that the Tesla driver was installed during the private image creation. If not, install the driver for computing acceleration after the ECS is created. For details, see Installing a Tesla Driver and CUDA Toolkit on a GPU-accelerated ECS.

Computing-accelerated P2v

Overview

P2v ECSs use NVIDIA Tesla V100 GPUs and deliver high flexibility, high-performance computing, and high cost-effectiveness. These ECSs use GPU NVLink for direct communication between GPUs, improving data transmission efficiency. P2v ECSs provide outstanding general computing capabilities and have strengths in AI-based deep learning, scientific computing, Computational Fluid Dynamics (CFD), computing finance, seismic analysis, molecular modeling, and genomics.

Specifications

Table 5 P2v ECS specifications

Flavor

vCPUs

Memory (GiB)

Maximum/Assured Bandwidth (Gbit/s)

Max. PPS (10,000)

NIC Multi-Queue

Maximum NICs

GPUs

GPU Connection

GPU Memory (GiB)

Virtualization Type

p2v.2xlarge.8

8

64

10/4

50

4

4

1 x V100

N/A

1 × 16 GiB

KVM

p2v.4xlarge.8

16

128

15/8

100

8

8

2 x V100

NVLink

2 × 16 GiB

KVM

p2v.8xlarge.8

32

256

25/15

200

16

8

4 x V100

NVLink

4 × 16 GiB

KVM

p2v.16xlarge.8

64

512

30/30

400

32

8

8 x V100

NVLink

8 × 16 GiB

KVM

P2v ECS Features
  • CPU: Intel® Xeon® Scalable 6151 processors (3.0 GHz of base frequency and 3.4 GHz of turbo frequency)
  • Up to eight NVIDIA Tesla V100 GPUs on an ECS
  • NVIDIA CUDA parallel computing and common deep learning frameworks, such as TensorFlow, Caffe, PyTorch, and MXNet
  • 15.7 TFLOPS of single-precision computing and 7.8 TFLOPS of double-precision computing
  • NVIDIA Tensor cores with 125 TFLOPS of single- and double-precision computing for deep learning
  • Up to 30 Gbit/s of network bandwidth on a single ECS
  • 16 GiB of HBM2 GPU memory with a bandwidth of 900 Gbit/s
  • Comprehensive basic capabilities

    Networks are user-defined, subnets can be divided, and network access policies can be configured as needed. Mass storage is used, elastic capacity expansion as well as backup and restoration are supported to make data more secure. Auto Scaling allows you to add or reduce the number of ECSs quickly.

  • Flexibility

    Similar to other types of ECSs, P2v ECSs can be provisioned in a few minutes.

  • Excellent supercomputing ecosystem

    The supercomputing ecosystem allows you to build up a flexible, high-performance, cost-effective computing platform. A large number of HPC applications and deep-learning frameworks can run on P2v ECSs.

Supported Common Software

P2v ECSs are used in computing acceleration scenarios, such as deep learning training, inference, scientific computing, molecular modeling, and seismic analysis. If the software is required to support GPU CUDA, use P2v ECSs. P2v ECSs support the following commonly used software:
  • Common deep learning frameworks, such as TensorFlow, Caffe, PyTorch, and MXNet
  • CUDA GPU rendering supported by RedShift for Autodesk 3dsMax and V-Ray for 3ds Max
  • Agisoft PhotoScan
  • MapD
Notes
  • P2v ECSs support the following OSs:
    • Windows Server 2016 Standard 64bit
    • Windows Server 2012 R2 Standard 64bit
    • Ubuntu Server 16.04 64bit
    • CentOS 7.7 64bit
    • EulerOS 2.5 64bit
  • By default, P2v ECSs created using a public image have the Tesla driver installed.
  • If a P2v ECS is created using a private image, make sure that the Tesla driver was installed during the private image creation. If not, install the driver for computing acceleration after the ECS is created. For details, see Installing a Tesla Driver and CUDA Toolkit on a GPU-accelerated ECS.

Inference-accelerated PI2

Overview

PI2 ECSs use NVIDIA Tesla T4 GPUs dedicated for real-time AI inference. These ECSs use the T4 INT8 calculator for up to 130 TOPS of INT8 computing. The PI2 ECSs can also be used for light-load training.

Specifications

Table 6 PI2 ECS specifications

Flavor

vCPUs

Memory (GiB)

Maximum/Assured Bandwidth (Gbit/s)

Max. PPS (10,000)

NIC Multi-Queue

GPUs

GPU Memory (GiB)

Local Disks

Virtualization Type

pi2.2xlarge.4

8

32

10/4

50

4

1 x T4

1×16

N/A

KVM

pi2.4xlarge.4

16

64

15/8

100

8

2 x T4

2×16

N/A

KVM

pi2.8xlarge.4

32

128

25/15

200

16

4 x T4

4×16

N/A

KVM

PI2 ECS Features

  • CPU: 2nd Generation Intel® Xeon® Scalable 6278 processors (2.6 GHz of base frequency and 3.5 GHz of turbo frequency), or Intel® Xeon® Scalable 6151 processors (3.0 GHz of base frequency and 3.4 GHz of turbo frequency)
  • Up to four NVIDIA Tesla T4 GPUs on an ECS
  • GPU hardware passthrough
  • Up to 8.1 TFLOPS of single-precision computing on a single GPU
  • Up to 130 TOPS of INT8 computing on a single GPU
  • 16 GiB of GDDR6 GPU memory with a bandwidth of 320 GiB/s on a single GPU
  • One built-in NVENC and two NVDEC GPUs

Supported Common Software

PI2 ECSs are used in GPU-based inference computing scenarios, such as image recognition, speech recognition, and natural language processing. The PI2 ECSs can also be used for light-load training.

PI2 ECSs support the following commonly used software:

  • Deep learning frameworks, such as TensorFlow, Caffe, PyTorch, and MXNet

Notes

  • The basic resources, including vCPUs, memory, and image of a pay-per-use PI2 ECS of flavor pi2.2xlarge.4, pi2.4xlarge.4, or pi2.8xlarge.4 are not billed after the ECS is stopped, but the system disk of the ECS is still being billed according to the disk capacity. The resources associated with the ECS, such as EVS disks, EIP, and bandwidth, are separately billed.

    The resources of a pay-per-use PI2 ECS of flavor pi2.2xlarge.4, pi2.4xlarge.4, or pi2.8xlarge.4 are released after the ECS is stopped. If the backend resources are insufficient when the ECS is started, starting the ECS may fail. If you want to use such an ECS for a long period of time, change its billing mode to yearly/monthly or do not stop the ECS.

  • PI2 ECSs support the following OSs:
    • Windows Server 2016 Standard 64bit
    • Ubuntu Server 16.04 64bit
    • CentOS 7.5 64bit
  • PI2 ECSs support automatic recovery when the hosts accommodating such ECSs become faulty.
  • By default, PI2 ECSs created using a public image have the Tesla driver installed.
  • If a PI2 ECS is created using a private image, make sure that the Tesla driver was installed during the private image creation. If not, install the driver for computing acceleration after the ECS is created. For details, see Installing a Tesla Driver and CUDA Toolkit on a GPU-accelerated ECS.

Inference-accelerated PI1

Overview

PI1 ECSs use NVIDIA Tesla P4 GPUs dedicated for real-time AI inference. Working with P4 INT8 calculators, PI1 ECSs have shortened the inference latency by 15 times. Working with hardware decoding engines, PI1 ECSs concurrently support real-time 35-channel HD video transcoding and inference.

Specifications

Table 7 PI1 ECS specifications

Flavor

vCPUs

Memory (GiB)

Maximum/Assured Bandwidth (Gbit/s)

Max. PPS (10,000)

NIC Multi-Queue

GPUs

GPU Memory (GiB)

Local Disks

Virtualization Type

pi1.2xlarge.4

8

32

5/1.6

40

2

1 x P4

1 × 8 GiB

N/A

KVM

pi1.4xlarge.4

16

64

8/3.2

70

4

2 x P4

2 × 8 GiB

N/A

KVM

pi1.8xlarge.4

32

128

10/6.5

140

8

4 x P4

4 × 8 GiB

N/A

KVM

PI1 ECS Features
  • CPU: Intel® Xeon® Processor E5-2697 v4 (2.3 GHz)
  • Up to four NVIDIA Tesla P4 GPUs on an ECS
  • GPU hardware passthrough
  • Up to 5.5 TFLOPS of single-precision computing on a single GPU
  • Up to 22 TOPS of INT8 computing on a single GPU
  • 8 GiB of ECC GPU memory with a bandwidth of 192 GiB/s on a single GPU
  • Hardware video encoding and decoding engines embedded in GPUs for concurrent real-time 35-channel HD video transcoding and inference

Supported Common Software

PI1 ECSs are used in GPU-based inference computing scenarios, such as image recognition, speech recognition, and natural language processing.

PI1 ECSs support the following commonly used software:

  • Deep learning frameworks, such as TensorFlow, Caffe, PyTorch, and MXNet
Notes
  • The basic resources, including vCPUs, memory, and image of a pay-per-use PI1 ECS of flavor pi1.2xlarge.4, pi1.4xlarge.4, or pi1.8xlarge.4 are not billed after the ECS is stopped, but the system disk of the ECS is still being billed according to the disk capacity. The resources associated with the ECS, such as EVS disks, EIP, and bandwidth, are separately billed.

    The resources of a pay-per-use PI1 ECS of flavor pi1.2xlarge.4, pi1.4xlarge.4, or pi1.8xlarge.4 are released after the ECS is stopped. If the backend resources are insufficient when the ECS is started, starting the ECS may fail. If you want to use such an ECS for a long period of time, change its billing mode to yearly/monthly or do not stop the ECS.

  • PI1 ECSs do not support specifications modification.
  • PI1 ECSs support the following OSs:
    • Ubuntu Server 16.04 64bit
    • Ubuntu Server 14.04 64bit
    • CentOS 7.3 64bit
  • PI1 ECSs support automatic recovery when the hosts accommodating such ECSs become faulty.
  • By default, PI1 ECSs created using a public image have the Tesla driver installed.
  • If a PI1 ECS is created using a private image, make sure that the Tesla driver was installed during the private image creation. If not, install the driver for computing acceleration after the ECS is created. For details, see Installing a Tesla Driver and CUDA Toolkit on a GPU-accelerated ECS.
close