This topic describes the features of GPU-accelerated compute-optimized and vGPU-accelerated instance families of Elastic Compute Service (ECS) and lists the instance types of each instance family.

sgn7i-vws, vGPU-accelerated instance family with shared CPUs

Features:
  • This instance family uses the third-generation SHENLONG architecture to provide predictable and consistent ultra-high performance. This instance family utilizes fast path acceleration on chips to improve storage performance, network performance, and computing stability by an order of magnitude. This way, data storage and model loading can be performed more quickly.
  • Instances of the sgn7i-vws instance family share CPU and network resources to maximize the utilization of underlying resources. Each instance has exclusive access to its memory and GPU memory to ensure data isolation and high performance.
    Note If you want to use exclusive CPU resources, select the vgn7i-vws instance family.
  • This instance family comes with a NVIDIA GRID vWS license and provides certified graphics acceleration capabilities for Computer Aided Design (CAD) software to meet the requirements of professional graphic design. Instances of this instance family can serve as lightweight GPU-accelerated compute-optimized instances to reduce the costs of small-scale AI inference tasks.
  • Compute:
    • Uses NVIDIA A10 GPUs that have the following features:
      • Innovative NVIDIA Ampere architecture
      • Support for acceleration features (such as vGPU, RTX, and TensorRT) to provide diversified business support
    • Uses 2.9 GHz Intel® Xeon® Scalable (Ice Lake) processors that deliver an all-core turbo frequency of 3.5 GHz.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports only enhanced SSDs (ESSDs).
      Note For more information about the performance of cloud disks, see EBS performance.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Concurrent AI inference tasks that require high-performance CPUs, memory, and GPUs, such as image recognition, speech recognition, and behavior identification
    • Compute-intensive graphics processing tasks that require high-performance 3D graphics virtualization capabilities, such as remote graphic design and cloud gaming
    • 3D modeling in fields that require the use of Ice Lake processors, such as animation and film production, cloud gaming, and mechanical design
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork baseline/burst bandwidth (Gbit/s)Packet forwarding rate (pps)Network interface controller (NIC) queuesElastic network interfaces (ENIs)
ecs.sgn7i-vws-m2.xlarge415.5NVIDIA A10 * 1/1224GB * 1/121.5/5500,00042
ecs.sgn7i-vws-m4.2xlarge831NVIDIA A10 * 1/624GB * 1/62.5/101,000,00044
ecs.sgn7i-vws-m8.4xlarge1662NVIDIA A10 * 1/324GB * 1/35/202,000,00084
ecs.sgn7i-vws-m2s.xlarge48NVIDIA A10 * 1/1224GB * 1/121.5/5500,00042
ecs.sgn7i-vws-m4s.2xlarge816NVIDIA A10 * 1/624GB * 1/62.5/101,000,00044
ecs.sgn7i-vws-m8s.4xlarge1632NVIDIA A10 * 1/324GB * 1/35/202,000,00084
Note
  • The GPU column in the preceding table indicates the GPU model and GPU slicing information for each instance type. Each GPU can be sliced into multiple GPU partitions, and each GPU partition can be assigned as a vGPU to an instance. Example:

    NVIDIA A10 * 1/12. NVIDIA A10 is the GPU model. 1/12 indicates that a GPU is sliced into 12 GPU partitions, and each GPU partition is assigned as a vGPU to an instance.

  • You can go to the ECS Instance Types Available for Each Region page to view the instance types available in each region.
  • For more information about these specifications, see the "Instance type specifications" section in Instance family.

vgn7i-vws, vGPU-accelerated instance family

Features:
  • This instance family uses the third-generation SHENLONG architecture to provide predictable and consistent ultra-high performance. This instance family utilizes fast path acceleration on chips to improve storage performance, network performance, and computing stability by an order of magnitude. This way, data storage and model loading can be performed more quickly.
  • This instance family comes with a NVIDIA GRID vWS license and provides certified graphics acceleration capabilities for CAD software to meet the requirements of professional graphic design. Instances of this instance family can serve as lightweight GPU-accelerated compute-optimized instances to reduce the costs of small-scale AI inference tasks.
  • Compute:
    • Uses NVIDIA A10 GPUs that have the following features:
      • Innovative NVIDIA Ampere architecture
      • Support for acceleration features (such as vGPU, RTX, and TensorRT) to provide diversified business support
    • Uses 2.9 GHz Intel® Xeon® Scalable (Ice Lake) processors that deliver an all-core turbo frequency of 3.5 GHz.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports only ESSDs.
      Note For more information about the performance of cloud disks, see EBS performance.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Concurrent AI inference tasks that require high-performance CPUs, memory, and GPUs, such as image recognition, speech recognition, and behavior identification
    • Compute-intensive graphics processing tasks that require high-performance 3D graphics virtualization capabilities, such as remote graphic design and cloud gaming
    • 3D modeling in fields that require the use of Ice Lake processors, such as animation and film production, cloud gaming, and mechanical design
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)NIC queuesENIs
ecs.vgn7i-vws-m4.xlarge430NVIDIA A10 * 1/624GB * 1/631,000,00044
ecs.vgn7i-vws-m8.2xlarge1062NVIDIA A10 * 1/324GB * 1/352,000,00086
ecs.vgn7i-vws-m12.3xlarge1493NVIDIA A10 * 1/224GB * 1/283,000,00086
ecs.vgn7i-vws-m24.7xlarge30186NVIDIA A10 * 124GB * 1166,000,000128
Note
  • The GPU column in the preceding table indicates the GPU model and GPU slicing information for each instance type. Each GPU can be sliced into multiple GPU partitions, and each GPU partition can be assigned as a vGPU to an instance. Example:

    NVIDIA A10 * 1/6. NVIDIA A10 is the GPU model. 1/6 indicates that a GPU is sliced into six GPU partitions, and each GPU partition is assigned as a vGPU to an instance.

  • You can go to the ECS Instance Types Available for Each Region page to view the instance types available in each region.
  • For more information about these specifications, see the "Instance type specifications" section in Instance family.

gn7s, GPU-accelerated compute-optimized instance family

Features:
  • This instance family uses new Intel Ice lake processors and NVIDIA A30 GPUs that are based on NVIDIA Ampere architecture. You can choose an appropriate mix of GPUs and CPU resources to meet various requirements for AI business.
  • This instance family uses third-generation SHENLONG architecture and doubles the average bandwidths of virtual private clouds (VPCs), networks, and disks compared with instance families of the previous generation.
  • Compute:
    • Uses NVIDIA A30 GPUs that have the following features:
      • Innovative NVIDIA Ampere architecture
      • Support for the multi-instance GPU (MIG) feature and acceleration features (based on second-generation Tensor cores) to provide diversified business support
    • Uses 2.9 GHz Intel® Xeon® Scalable (Ice Lake) processors that deliver an all-core turbo frequency of 3.5 GHz.
    • Improves memory sizes significantly from instance families of the previous generation.
  • Storage: Supports only ESSDs.
    Note For more information about the performance of cloud disks, see EBS performance.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios: concurrent AI inference tasks that require high-performance CPUs, memory, and GPUs, such as image recognition, speech recognition, and behavior identification.
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)IPv6 addresses per ENINIC queuesENIs
ecs.gn7s-c8g1.2xlarge860NVIDIA A30 * 124GB * 1166,000,0001128
ecs.gn7s-c16g1.4xlarge16120NVIDIA A30 * 124GB * 1166,000,0001128
ecs.gn7s-c32g1.8xlarge32250NVIDIA A30 * 124GB * 1166,000,0001128
ecs.gn7s-c32g1.16xlarge64500NVIDIA A30 * 224GB * 23212,000,00011615
ecs.gn7s-c32g1.32xlarge1281000NVIDIA A30 * 424GB * 46424,000,00013215
ecs.gn7s-c48g1.12xlarge48380NVIDIA A30 * 124GB * 1166,000,0001128
ecs.gn7s-c56g1.14xlarge56440NVIDIA A30 * 124GB * 1166,000,0001128
Note

gn7e, GPU-accelerated compute-optimized instance family

Features:
  • This instance family uses Intel Ice lake processors and NVIDIA A100 SXM4 80GB GPUs that are interconnected by using NVSwitches. You can choose an appropriate mix of GPUs and CPU resources to meet various AI business requirements.
  • This instance family uses the third-generation SHENLONG architecture and doubles the average bandwidths of VPCs, networks, and disks compared with instance families of the previous generation.
  • Compute:
    • Uses NVIDIA A100 GPUs that have the following features:
      • Innovative NVIDIA Ampere architecture
      • Connections established between NVIDIA A100 GPUs by using NVSwitches
      • Support for mixed-precision computing with 80 GB of HBM2 memory per GPU
    • Uses 2.9 GHz Intel® Xeon® Platinum 8369B (Ice Lake) processors that deliver an all-core turbo frequency of 3.5 GHz.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports only ESSDs.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Small- and medium-scale AI training
    • High-performance computing (HPC) business accelerated by using Compute Unified Device Architecture (CUDA)
    • AI inference tasks that require high GPU processing capabilities or large amounts of GPU memory
    • Deep learning applications such as training applications of AI algorithms used in image classification, autonomous vehicles, and speech recognition
    • Scientific computing applications that require robust GPU computing capabilities such as computational fluid dynamics, computational finance, molecular dynamics, and environmental analytics.
    Important When you use AI training services that feature a high communication load, such as transformer models, you must enable NVlink for GPU-to-GPU communication. Otherwise, data may be damaged due to unpredictable failures that are caused by large-scale data transmission over Peripheral Component Interconnect Express (PCIe) links. If you do not understand the topology of the communication links that are used for AI training services, submit a ticket to obtain technical support.
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)NIC queuesENIsPrivate IP addresses per ENI
ecs.gn7e-c16g1.4xlarge16125NVIDIA A100 * 180GB * 183,000,0008810
ecs.gn7e-c16g1.16xlarge64500NVIDIA A100 * 480GB * 43212,000,00032810
ecs.gn7e-c16g1.32xlarge1281000NVIDIA A100 * 880GB * 86424,000,000321615
Note

gn7i, GPU-accelerated compute-optimized instance family

Features:
  • This instance family uses the third-generation SHENLONG architecture to provide predictable and consistent ultra-high performance. This instance family utilizes fast path acceleration on chips to improve storage performance, network performance, and computing stability by an order of magnitude.
  • Compute:
    • Uses NVIDIA A10 GPUs that have the following features:
      • Innovative NVIDIA Ampere architecture
      • Support for acceleration features such as RTX and TensorRT
    • Uses 2.9 GHz Intel® Xeon® Scalable (Ice Lake) processors that deliver an all-core turbo frequency of 3.5 GHz.
    • Provides memory of up to 752 GiB, which is much larger than the memory sizes of the gn6i instance family.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports only ESSDs.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Concurrent AI inference tasks that require high-performance CPUs, memory, and GPUs, such as image recognition, speech recognition, and behavior identification
    • Compute-intensive graphics processing tasks that require high-performance 3D graphics virtualization capabilities, such as remote graphic design and cloud gaming.
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)NIC queuesENIsPrivate IP addresses per ENI
ecs.gn7i-c8g1.2xlarge830NVIDIA A10 * 124GB * 1161,600,000845
ecs.gn7i-c16g1.4xlarge1660NVIDIA A10 * 124GB * 1163,000,000885
ecs.gn7i-c32g1.8xlarge32188NVIDIA A10 * 124GB * 1166,000,0001285
ecs.gn7i-c32g1.16xlarge64376NVIDIA A10 * 224GB * 23212,000,00016155
ecs.gn7i-c32g1.32xlarge128752NVIDIA A10 * 424GB * 46424,000,000321510
ecs.gn7i-c48g1.12xlarge48310NVIDIA A10 * 124GB * 1169,000,0001688
ecs.gn7i-c56g1.14xlarge56346NVIDIA A10 * 124GB * 11612,000,00016128
Note

gn7, GPU-accelerated compute-optimized instance family

Features:
  • Compute:
    • Uses NVIDIA A100 GPUs. NVSwitches are used to establish connections between NVIDIA A100 GPUs. The GPUs have the following features:
      • Innovative NVIDIA Ampere architecture
      • 40 GB of HBM2 memory per GPU
    • Uses 2.5 GHz Intel® Xeon® Platinum 8269CY (Cascade Lake) processors.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports only ESSDs.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Deep learning applications such as training applications of AI algorithms used in image classification, autonomous vehicles, and speech recognition
    • Scientific computing applications that require robust GPU computing capabilities such as computational fluid dynamics, computational finance, molecular dynamics, and environmental analytics
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)NIC queuesENIs
ecs.gn7-c12g1.3xlarge1294NVIDIA A100 * 140GB * 142,500,00048
ecs.gn7-c13g1.13xlarge52378NVIDIA A100 * 440GB * 4169,000,000168
ecs.gn7-c13g1.26xlarge104756NVIDIA A100 * 840GB * 83018,000,0001615
Note

vgn6i and vgn6i-vws, vGPU-accelerated instance families

vgn6i-vws:
  • In light of the NVIDIA GRID driver upgrade, Alibaba Cloud upgrades the vgn6i instance family to vgn6i-vws instance family. The vgn6i-vws instance family uses the latest NVIDIA GRID driver and provides a NVIDIA GRID vWS license. Submit a ticket to apply for free images that have the NVIDIA GRID driver pre-installed.
  • To use other public images or custom images that do not contain a NVIDIA GRID driver, submit a ticket to apply for the GRID driver file and install the NVIDIA GRID driver separately. Alibaba Cloud does not charge additional fees for the license of the GRID driver.
vgn6i:
  • If you want your vgn6i instance to support graphics features such as Open Graphics Library (OpenGL), you must purchase a GRID license from NVIDIA. After the instance is created, manually install a GRID driver and activate the license.
  • Compute:
    • Uses NVIDIA T4 GPUs.
    • Uses vGPUs.
      • Supports the 1/4 and 1/2 computing capacity of NVIDIA Tesla T4 GPUs.
      • Supports 4 GB and 8 GB of GPU memory.
    • Offers a CPU-to-memory ratio of 1:5.
    • Uses 2.5 GHz Intel® Xeon® Platinum 8163 (Skylake) processors.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports only standard SSDs and ultra disks.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Real-time rendering for cloud gaming
    • Real-time rendering for AR and VR applications
    • AI (deep learning and machine learning) inference for elastic Internet service deployment
    • Educational environment of deep learning
    • Modeling experiment environment of deep learning
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)NIC queues (primary ENI/secondary ENI)ENIsPrivate IP addresses per ENI
ecs.vgn6i-m4.xlarge423NVIDIA T4 * 1/416GB * 1/42500,0004/2310
ecs.vgn6i-m8.2xlarge1046NVIDIA T4 * 1/216GB * 1/24800,0008/2410
ecs.vgn6i-m4-vws.xlarge423NVIDIA T4 * 1/416GB * 1/42500,0004/2310
ecs.vgn6i-m8-vws.2xlarge1046NVIDIA T4 * 1/216GB * 1/24800,0008/2410
ecs.vgn6i-m16-vws.5xlarge2092NVIDIA T4 * 116GB * 17.51,200,0006410
Note
  • The GPU column in the preceding table indicates the GPU model and GPU slicing information for each instance type. Each GPU can be sliced into multiple GPU partitions, and each GPU partition can be assigned as a vGPU to an instance. Example:

    NVIDIA T4 GPU * 1/4. NVIDIA T4 is the GPU model. 1/4 indicates that a GPU is sliced into four GPU partitions, and each GPU partition is assigned as a vGPU to an instance.

  • You can go to the ECS Instance Types Available for Each Region page to view the instance types available in each region.
  • For more information about these specifications, see the "Instance type specifications" section in Instance family.

gn6i, GPU-accelerated compute-optimized instance family

Features:
  • Compute:
    • Uses NVIDIA T4 GPUs that have the following features:
      • Innovative NVIDIA Turing architecture
      • 16 GB memory (320 GB/s bandwidth) per GPU
      • 2,560 CUDA cores per GPU
      • Up to 320 Turing Tensor cores per GPU
      • Mixed-precision Tensor cores that support 65 FP16 TFLOPS, 130 INT8 TOPS, and 260 INT4 TOPS
    • Offers a CPU-to-memory ratio of 1:4.
    • Uses 2.5 GHz Intel Xeon Platinum 8163 (Skylake) processors.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports standard SSDs, ultra disks, and ESSDs that deliver millions of IOPS.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • AI (deep learning and machine learning) inference for computer vision, speech recognition, speech synthesis, natural language processing (NLP), machine translation, and recommendation systems
    • Real-time rendering for cloud games
    • Real-time rendering for AR and VR applications
    • Graphics workstations or overloaded graphics computing
    • GPU-accelerated databases
    • High-performance computing
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)Baseline storage IOPSNIC queuesENIsPrivate IP addresses per ENI
ecs.gn6i-c4g1.xlarge415NVIDIA T4 * 116GB * 14500,000N/A2210
ecs.gn6i-c8g1.2xlarge831NVIDIA T4 * 116GB * 15800,000N/A2210
ecs.gn6i-c16g1.4xlarge1662NVIDIA T4 * 116GB * 161,000,000N/A4310
ecs.gn6i-c24g1.6xlarge2493NVIDIA T4 * 116GB * 17.51,200,000N/A6410
ecs.gn6i-c40g1.10xlarge40155NVIDIA T4 * 116GB * 1101,600,000N/A161010
ecs.gn6i-c24g1.12xlarge48186NVIDIA T4 * 216GB * 2152,400,000N/A12610
ecs.gn6i-c24g1.24xlarge96372NVIDIA T4 * 416GB * 4304,800,000250,00024810
Note

gn6e, GPU-accelerated compute-optimized instance family

Features:
  • Compute:
    • Uses NVIDIA V100 GPUs that each has 32 GB of GPU memory and support NVLink.
    • Uses NVIDIA V100 GPUs (SXM2-based) that have the following features:
      • Innovative NVIDIA Volta architecture
      • 32 GB of HBM2 memory (900 GB/s bandwidth) per GPU
      • 5,120 CUDA cores per GPU
      • 640 Tensor cores per GPU
      • Support for up to six NVLink bidirectional connections, which each provide a bandwidth of 25 GB/s in each direction for a total bandwidth of 300 GB/s (6 × 25 × 2 = 300)
    • Offers a CPU-to-memory ratio of 1:8.
    • Uses 2.5 GHz Intel® Xeon® Platinum 8163 (Skylake) processors.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports ESSDs, standard SSDs, and ultra disks.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Deep learning applications such as the training and inference applications of AI algorithms used in image classification, autonomous driving, and speech recognition
    • Scientific computing applications, such as computational fluid dynamics, computational finance, molecular dynamics, and environmental analytics.
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)NIC queuesENIsPrivate IP addresses per ENI
ecs.gn6e-c12g1.3xlarge1292NVIDIA V100 * 132GB * 15800,0008610
ecs.gn6e-c12g1.12xlarge48368NVIDIA V100 * 432GB * 4162,400,0008820
ecs.gn6e-c12g1.24xlarge96736NVIDIA V100 * 832GB * 8324,800,00016820
Note

gn6v, GPU-accelerated compute-optimized instance family

Features:
  • Compute:
    • Uses NVIDIA V100 GPUs.
    • Uses NVIDIA V100 GPUs (SXM2-based) that have the following features:
      • Innovative NVIDIA Volta architecture
      • 16 GB of HBM2 memory (900 GB/s bandwidth) per GPU
      • 5,120 CUDA cores per GPU
      • 640 Tensor cores per GPU
      • Support for up to six NVLink bidirectional connections, which each provide a bandwidth of 25 GB/s in each direction for a total bandwidth of 300 GB/s (6 × 25 × 2 = 300)
    • Offers a CPU-to-memory ratio of 1:4.
    • Uses 2.5 GHz Intel® Xeon® Platinum 8163 (Skylake) processors.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports ESSDs, standard SSDs, and ultra disks.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Deep learning applications such as the training and inference applications of AI algorithms used in image classification, autonomous driving, and speech recognition
    • Scientific computing applications, such as computational fluid dynamics, computational finance, molecular dynamics, and environmental analytics.
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)Baseline storage IOPSNIC queuesENIsPrivate IP addresses per ENI
ecs.gn6v-c8g1.2xlarge832NVIDIA V100 * 116GB * 12.5800,000N/A4410
ecs.gn6v-c8g1.8xlarge32128NVIDIA V100 * 416GB * 4102,000,000N/A8820
ecs.gn6v-c8g1.16xlarge64256NVIDIA V100 * 816GB * 8202,500,000N/A16820
ecs.gn6v-c10g1.20xlarge82336NVIDIA V100 * 816GB * 8324,500,000250,00016820
Note

vgn5i, vGPU-accelerated instance family

Features:
  • If you want your vgn5i instance to support graphics features such as OpenGL, you must purchase a GRID license from NVIDIA. After the instance is created, manually install a GRID driver and activate the license.
  • Compute:
    • Uses NVIDIA P4 GPUs.
    • Uses vGPUs.
      • Supports the 1/8, 1/4, 1/2, and 1/1 computing capacity of NVIDIA Tesla P4 GPUs.
      • Supports 1 GB, 2 GB, 4 GB, and 8 GB of GPU memory.
    • Offers a CPU-to-memory ratio of 1:3.
    • Uses 2.5 GHz Intel® Xeon® E5-2682 v4 (Broadwell) processors.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports standard SSDs and ultra disks.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Real-time rendering for cloud gaming
    • Real-time rendering for AR and VR applications
    • AI (deep learning and machine learning) inference for elastic Internet service deployment
    • Educational environment of deep learning
    • Modeling experiment environment of deep learning
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)NIC queuesENIsPrivate IP addresses per ENI
ecs.vgn5i-m1.large26NVIDIA P4 * 1/88GB * 1/81300,000226
ecs.vgn5i-m2.xlarge412NVIDIA P4 * 1/48GB * 1/42500,0002310
ecs.vgn5i-m4.2xlarge824NVIDIA P4 * 1/28GB * 1/23800,0002410
ecs.vgn5i-m8.4xlarge1648NVIDIA P4 * 18GB * 151,000,0004520
Note
  • The GPU column in the preceding table indicates the GPU model and GPU slicing information for each instance type. Each GPU can be sliced into multiple GPU partitions, and each GPU partition can be assigned as a vGPU to an instance. Example:

    NVIDIA P4 * 1/8. NVIDIA P4 is the GPU model. 1/8 indicates that a GPU is sliced into eight GPU partitions, and each GPU partition is assigned as a vGPU to an instance.

  • You can go to the ECS Instance Types Available for Each Region page to view the instance types available in each region.
  • For more information about these specifications, see the "Instance type specifications" section in Instance family.

gn5, GPU-accelerated compute-optimized instance family

Features:
  • Compute:
    • Uses NVIDIA P100 GPUs.
    • Offers multiple CPU-to-memory ratios.
    • Uses 2.5 GHz Intel® Xeon®E5-2682 v4 (Broadwell) processors.
  • Storage:
    • Supports high-performance local Non-Volatile Memory Express (NVMe) SSDs.
    • Is an instance family in which all instances are I/O optimized.
    • Supports standard SSDs and ultra disks.
  • Network:
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Deep learning
    • Scientific computing applications, such as computational fluid dynamics, computational finance, genomics, and environmental analytics
    • Server-side GPU compute workloads such as high-performance computing, rendering, and multi-media encoding and decoding
Instance types
Instance typevCPUsMemory (GiB)Local storage (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)NIC queuesENIsPrivate IP addresses per ENI
ecs.gn5-c4g1.xlarge430440NVIDIA P100 * 116GB * 13300,0001310
ecs.gn5-c8g1.2xlarge860440NVIDIA P100 * 116GB * 13400,0001410
ecs.gn5-c4g1.2xlarge860880NVIDIA P100 * 216GB * 251,000,0002410
ecs.gn5-c8g1.4xlarge16120880NVIDIA P100 * 216GB * 251,000,0004820
ecs.gn5-c28g1.7xlarge28112440NVIDIA P100 * 116GB * 151,000,0008820
ecs.gn5-c8g1.8xlarge322401760NVIDIA P100 * 416GB * 4102,000,0008820
ecs.gn5-c28g1.14xlarge56224880NVIDIA P100 * 216GB * 2102,000,00014820
ecs.gn5-c8g1.14xlarge544803520NVIDIA P100 * 816GB * 8254,000,00014820
Note

gn5i, GPU-accelerated compute-optimized instance family

Features:
  • Compute:
    • Uses NVIDIA P4 GPUs.
    • Offers a CPU-to-memory ratio of 1:4.
    • Uses 2.5 GHz Intel® Xeon®E5-2682 v4 (Broadwell) processors.
  • Storage:
    • Is an instance family in which all instances are I/O optimized.
    • Supports standard SSDs and ultra disks.
  • Network:
    • Supports IPv6.
    • Provides high network performance based on large computing capacity.
  • Supported scenarios:
    • Deep learning inference
    • Server-side GPU compute workloads such as multi-media encoding and decoding
Instance types
Instance typevCPUsMemory (GiB)GPUGPU memoryNetwork bandwidth (Gbit/s)Packet forwarding rate (pps)NIC queuesENIsPrivate IP addresses per ENI
ecs.gn5i-c2g1.large28NVIDIA P4 * 18GB * 11100,000226
ecs.gn5i-c4g1.xlarge416NVIDIA P4 * 18GB * 11.5200,0002310
ecs.gn5i-c8g1.2xlarge832NVIDIA P4 * 18GB * 12400,0004410
ecs.gn5i-c16g1.4xlarge1664NVIDIA P4 * 18GB * 13800,0004820
ecs.gn5i-c16g1.8xlarge32128NVIDIA P4 * 28GB * 261,200,0008820
ecs.gn5i-c28g1.14xlarge56224NVIDIA P4 * 28GB * 2102,000,00014820
Note