This topic describes the details of all retired GPU-accelerated instance types. You can use recommended instance types based on your business requirements on use scenarios and computing capabilities.
For more information about available instance types, see Overview of instance families.
If you are using a retired instance type, we recommend that you change the instance type to another instance type that is available for purchase. For more information about the supported changes between instance types, see Instance types and families that support instance type changes.
ebmgn6ia, GPU-accelerated compute-optimized ECS Bare Metal Instance family
sccgn6e, GPU-accelerated compute-optimized SCC instance family
sccgn6, GPU-accelerated compute-optimized SCC instance family
sccgn6ne, GPU-accelerated compute-optimized SCC instance family
ebmgn6ia, GPU-accelerated compute-optimized ECS Bare Metal Instance family
Introduction:
This instance family uses the third-generation SHENLONG architecture and fast path acceleration on chips to provide predictable and consistent ultra-high computing, storage, and network performance.
This instance family uses NVIDIA T4 GPUs to offer GPU acceleration capabilities for graphics and AI applications and adopts container technology to start at least 60 virtual Android devices and provide hardware-accelerated video transcoding.
Supported scenarios:
Remote application services based on Android, such as always-on cloud-based services, cloud-based mobile games, cloud-based mobile phones, and Android service crawlers.
Compute:
Offers a CPU-to-memory ratio of 1:3.
Uses 2.8 GHz Ampere® Altra® Arm-based processors that deliver a turbo frequency of 3.0 GHz and provides high performance and high compatibility with applications for Android servers.
Storage:
Is an instance family in which all instances are I/O optimized.
Supports Enterprise SSDs (ESSDs) and ESSD AutoPL disks. For information about disks, see Overview of Block Storage.
Network:
Supports IPv4 and IPv6. For information about IPv6 communication, see IPv6 communication.
ebmgn6ia instance types
Instance type | vCPUs | Memory (GiB) | GPUs | GPU memory | Network baseline bandwidth (Gbit/s) | Packet forwarding rate (pps) | NIC queues | ENIs | Private IPv4 addresses per ENI | IPv6 addresses per ENI |
ecs.ebmgn6ia.20xlarge | 80 | 256 | NVIDIA T4 * 2 | 16GB * 2 | 32 | 24,000,000 | 32 | 15 | 10 | 1 |
Ampere® Altra® processors have specific requirements for operating system kernels. Instances of the preceding instance type can use Alibaba Cloud Linux 3 images and CentOS 8.4 or later images. We recommend that you use Alibaba Cloud Linux 3 images on the instances. If you want to use a different operating system distribution, patch the kernel of an instance that runs an operating system of that distribution, create a custom image from the instance, and then use the custom image to create instances of the instance type. For information about kernel patches, visit Ampere Altra (TM) Linux Kernel Porting Guide.
vgn6i, vGPU-accelerated instance family
Features:
Compute:
Uses NVIDIA T4 GPUs.
Uses vGPUs.
Supports the 1/4 and 1/2 compute 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 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 Augmented Reality (AR) and Virtual Reality (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
vgn6i instance types
Instance type | vCPUs | Memory (GiB) | GPU | GPU memory | Network baseline bandwidth (Gbit/s) | Packet forwarding rate (pps) | NIC queues (primary NIC/secondary NIC) | ENIs | Private IPv4 addresses per ENI |
ecs.vgn6i-m4.xlarge | 4 | 23 | NVIDIA T4 * 1/4 | 16GB * 1/4 | 2 | 500,000 | 4/2 | 3 | 10 |
ecs.vgn6i-m8.2xlarge | 10 | 46 | NVIDIA T4 * 1/2 | 16GB * 1/2 | 4 | 800,000 | 8/2 | 4 | 10 |
vgn5i, vGPU-accelerated instance family
Features:
Compute:
Uses NVIDIA P4 GPUs.
Uses vGPUs.
Supports the 1/8, 1/4, 1/2, and 1/1 compute 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
vgn5i instance types
Instance type | vCPUs | Memory (GiB) | GPUs | GPU memory | Network baseline bandwidth (Gbit/s) | Packet forwarding rate (pps) | NIC queues | ENIs | Private IPv4 addresses per ENI |
ecs.vgn5i-m1.large | 2 | 6 | NVIDIA P4 * 1/8 | 8GB * 1/8 | 1 | 300,000 | 2 | 2 | 6 |
ecs.vgn5i-m2.xlarge | 4 | 12 | NVIDIA P4 * 1/4 | 8GB * 1/4 | 2 | 500,000 | 2 | 3 | 10 |
ecs.vgn5i-m4.2xlarge | 8 | 24 | NVIDIA P4 * 1/2 | 8GB * 1/2 | 3 | 800,000 | 2 | 4 | 10 |
ecs.vgn5i-m8.4xlarge | 16 | 48 | NVIDIA P4 * 1 | 8GB * 1 | 5 | 1,000,000 | 4 | 5 | 20 |
The GPUs 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 allocated 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 can be allocated as a vGPU to an instance.
sccgn6e, GPU-accelerated compute-optimized SCC instance family
Features:
This instance family supports all features of ECS Bare Metal Instance. For more information, see ECS Bare Metal Instance families.
Compute:
Uses GPUs that have the following features:
Innovative Volta architecture
32 GB of HBM2 GPU memory
CUDA Cores 5120
Tensor Cores 640
GPU memory bandwidth of up to 900 GB/s
Support for up to six bidirectional NVLink connections, each of which has a unidirectional bandwidth of 25 GB/s for a total bandwidth of 300 GB/s
Offers a CPU-to-memory ratio of 1:8.
Uses 2.5 GHz Intel® Xeon® Platinum 8163 (Skylake) processors to provide consistent computing performance.
Storage:
Is an instance family in which all instances are I/O optimized.
Supports ESSDs, ESSD AutoPL disks, standard SSDs, and ultra disks.
Supports high-performance CPFS.
Network:
Supports IPv6.
Supports VPCs.
Supports RoCE v2 networks, which are dedicated to low-latency RDMA communication.
Supported scenarios:
Ultra-large-scale training for machine learning on distributed GPU clusters
Large-scale high-performance scientific computing and simulations
Large-scale data analytics, batch processing, and video encoding
sccgn6e instance types
Instance type | vCPUs | Memory (GiB) | GPUs | GPU memory (GB) | Network baseline bandwidth (Gbit/s) | Packet forwarding rate (pps) | RoCE network bandwidth (Gbit/s) | NIC queues | ENIs | Private IPv4 addresses per ENI |
ecs.sccgn6e.24xlarge | 96 | 768.0 | NVIDIA V100 * 8 | 32GB * 8 | 32 | 4,800,000 | 50 | 8 | 32 | 10 |
sccgn6, GPU-accelerated compute-optimized SCC instance family
Features:
This instance family supports all features of ECS Bare Metal Instance. For more information, see ECS Bare Metal Instance families.
Compute:
Uses NVIDIA V100 GPUs (SXM2-based) that have the following features:
Innovative Volta architecture
16 GB of HBM2 GPU memory
CUDA Cores 5120
Tensor Cores 640
GPU memory bandwidth of up to 900 GB/s
Support for up to six bidirectional NVLink connections, each of which has a unidirectional bandwidth of 25 GB/s for a total bandwidth of 300 GB/s
Offers a CPU-to-memory ratio of 1:4.
Uses 2.5 GHz Intel® Xeon® Platinum 8163 (Skylake) processors to provide consistent computing performance.
Storage:
Is an instance family in which all instances are I/O optimized.
Supports ESSDs, ESSD AutoPL disks, standard SSDs, and ultra disks.
Supports high-performance Cloud Paralleled File System (CPFS).
Network:
Supports IPv6.
Supports VPCs.
Supports RDMA over Converged Ethernet (RoCE) v2 networks, which are dedicated to low-latency remote direct memory access (RDMA) communication.
Supported scenarios:
Ultra-large-scale training for machine learning on distributed GPU clusters
Large-scale high-performance scientific computing and simulations
Large-scale data analytics, batch processing, and video encoding
sccgn6 instance types
Instance type | vCPUs | Memory (GiB) | GPUs | Network baseline bandwidth (Gbit/s) | Packet forwarding rate (pps) | RoCE network bandwidth (Gbit/s) | NIC queues | ENIs | Private IPv4 addresses per ENI |
ecs.sccgn6.24xlarge | 96 | 384.0 | NVIDIA V100 * 8 | 30 | 4,500,000 | 50 | 8 | 32 | 10 |
sccgn6ne, GPU-accelerated compute-optimized SCC instance family
Features:
This instance family provides all features of ECS Bare Metal Instance.
Compute:
Uses NVIDIA V100 GPUs (SXM2-based) that have the following features:
Innovative Volta architecture
32 GB of HBM2 GPU memory
CUDA Cores 5120
Tensor Cores 640
GPU memory bandwidth of up to 900 GB/s
Support for up to six bidirectional NVLink connections, each of which has a unidirectional bandwidth of 25 GB/s for a total bandwidth of 300 GB/s
Offers a CPU-to-memory ratio of 1:4.
Uses 2.5 GHz Intel® Xeon® Platinum 8163 (Skylake) processors to provide consistent computing performance.
Storage:
Is an instance family in which all instances are I/O optimized.
Supports ESSDs, standard SSDs, and ultra disks.
Supports high-performance CPFS.
Network:
Supports IPv6.
Supports VPCs.
Supports RoCE v2 networks, which are dedicated to low-latency RDMA communication.
Supported scenarios:
Ultra-large-scale training for machine learning on distributed GPU clusters
Large-scale high-performance scientific computing and simulations
Large-scale data analytics, batch processing, and video encoding
sccgn6ne instance types
Instance type | vCPUs | Memory (GiB) | GPUs | GPU memory | Network baseline bandwidth (Gbit/s) | Packet forwarding rate (pps) | RoCE network bandwidth (Gbit/s) | NIC queues | ENIs | Private IPv4 addresses per ENI |
ecs.sccgn6ne.24xlarge | 96 | 768.0 | NVIDIA V100 * 8 | 32GB * 8 | 32.0 | 4,800,000 | 100 | 16 | 8 | 20 |
gn4, GPU-accelerated compute-optimized instance family
Features:
Uses NVIDIA M40 GPUs.
Compute:
Offers multiple CPU-to-memory ratios.
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:
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
gn4 instance types
Instance type | vCPUs | Memory (GiB) | GPUs | GPU memory | Network baseline bandwidth (Gbit/s) | Packet forwarding rate (pps) | NIC queues | ENIs | Private IPv4 addresses per ENI |
ecs.gn4-c4g1.xlarge | 4 | 30.0 | NVIDIA M40 * 1 | 12GB * 1 | 3.0 | 300,000 | 1 | 3 | 10 |
ecs.gn4-c8g1.2xlarge | 8 | 30.0 | NVIDIA M40 * 1 | 12GB * 1 | 3.0 | 400,000 | 1 | 4 | 10 |
ecs.gn4.8xlarge | 32 | 48.0 | NVIDIA M40 * 1 | 12GB * 1 | 6.0 | 800,000 | 3 | 8 | 20 |
ecs.gn4-c4g1.2xlarge | 8 | 60.0 | NVIDIA M40 * 2 | 12GB * 2 | 5.0 | 500,000 | 1 | 4 | 10 |
ecs.gn4-c8g1.4xlarge | 16 | 60.0 | NVIDIA M40 * 2 | 12GB * 2 | 5.0 | 500,000 | 1 | 8 | 20 |
ecs.gn4.14xlarge | 56 | 96.0 | NVIDIA M40 * 2 | 12GB * 2 | 10.0 | 1,200,000 | 4 | 8 | 20 |
ga1, GPU-accelerated compute-optimized instance family
Features:
Uses AMD S7150 GPUs.
Supports high-performance local Non-Volatile Memory Express (NVMe) SSDs.
Compute:
Offers a CPU-to-memory ratio of 1:2.5.
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:
Provides high network performance based on large computing capacity.
Supported scenarios:
Rendering and multi-media encoding and decoding
Machine learning, high-performance computing, and high-performance databases
Server-side workloads that require powerful parallel floating-point computing capacity
ga1 instance types
Instance type | vCPUs | Memory (GiB) | Local storage (GiB) | GPUs | GPU memory | Network baseline bandwidth (Gbit/s) | Packet forwarding rate (pps) | NIC queues | ENIs | Private IPv4 addresses per ENI |
ecs.ga1.xlarge | 4 | 10.0 | 1 * 87 | AMD S7150 * 1/4 | 8GB * 1/4 | 1.0 | 200,000 | 1 | 3 | 10 |
ecs.ga1.2xlarge | 8 | 20.0 | 1 * 175 | AMD S7150 * 1/2 | 8GB * 1/2 | 1.5 | 300,000 | 1 | 4 | 10 |
ecs.ga1.4xlarge | 16 | 40.0 | 1 * 350 | AMD S7150 * 1 | 8GB * 1 | 3.0 | 500,000 | 2 | 8 | 20 |
ecs.ga1.8xlarge | 32 | 80.0 | 1 * 700 | AMD S7150 * 2 | 8GB * 2 | 6.0 | 800,000 | 3 | 8 | 20 |
ecs.ga1.14xlarge | 56 | 160.0 | 1 * 1400 | AMD S7150 * 4 | 8GB * 4 | 10.0 | 1,200,000 | 4 | 8 | 20 |