Computing-accelerated
|
P3
@@ -148,7 +155,7 @@
CUDA, DirectCompute, OpenACC, and OpenCLA single card is equipped with 10,752 CUDA cores, 84 second-generation RT cores, and 336 third-generation Tensor cores.Graphics applications acceleratedHeavy-load CPU inferenceApplication flow identical to common ECSsAutomatic scheduling of G7 ECSs to AZs where NVIDIA A40 GPUs are usedOne NVENC (encoding) engine and two NVDEC (decoding) engines (including AV1 decoding) embedded
Supported Common Software
G7 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 G7 ECSs. G7 ECSs support the following commonly used graphics processing software:
-- AutoCAD
- 3DS MAX
- MAYA
- Agisoft PhotoScan
- ContextCapture
- Adobe Premiere Pro
- Solidworks
- Unreal Engine
- Blender
- Vray
+- AutoCAD
- 3ds Max
- MAYA
- Agisoft PhotoScan
- ContextCapture
- Adobe Premiere Pro
- Solidworks
- Unreal Engine
- Blender
- Vray
Notes
- After a G7 ECS is stopped, basic resources (including vCPUs, memory, image, and GPUs) are not billed, but its system disk is billed based on the disk capacity. If other products, such as EVS disks, EIP, and bandwidth are associated with the ECS, these products are billed separately.
 Resources will be released after a G7 ECS is stopped. If resources are insufficient at the next start, the start may fail. If you want to use such an ECS for a long period of time, do not stop the ECS.
- CUDA and OpenCL
- NVIDIA T4 GPUs
- Graphics applications accelerated
- Heavy-load CPU inference
- Automatic scheduling of G6 ECSs to AZs where NVIDIA T4 GPUs are used
- One NVENC engine and two NVDEC engines embedded
Supported Common Software
G6 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 G6 ECSs. G6 ECSs support the following commonly used graphics processing software:
-- AutoCAD
- 3DS MAX
- MAYA
- Agisoft PhotoScan
- ContextCapture
+- AutoCAD
- 3ds Max
- MAYA
- Agisoft PhotoScan
- ContextCapture
Notes
- After a G6 ECS is stopped, basic resources (including vCPUs, memory, image, and GPUs) are not billed, but its system disk is billed based on the disk capacity. If other products, such as EVS disks, EIP, and bandwidth are associated with the ECS, these products are billed separately.
Resources will be released after a G6 ECS is stopped. If resources are insufficient at the next start, the start may fail. If you want to use such an ECS for a long period of time, do not stop the ECS.
@@ -266,117 +273,238 @@
- If a G6 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.
- GPU-accelerated ECSs differ greatly in general-purpose and heterogeneous computing power. Their specifications can only be changed to other specifications of the same instance type.
- GPU-accelerated ECSs do not support live migration.
+Computing-accelerated P5sOverview
+ P5s ECSs use high-performance NVIDIA Tesla H100 PCIe and provide outstanding real-time inference.
+ Specifications
+
+ Table 4 P5s ECS specificationsFlavor
+ |
+vCPUs
+ |
+Memory (GiB)
+ |
+Max./Assured Bandwidth (Gbit/s)
+ |
+Max. PPS (10,000)
+ |
+Max. NIC Queues
+ |
+Max. NICs
+ |
+GPUs
+ |
+GPU Memory (GiB)
+ |
+Virtualization
+ |
+
|---|
+
+p5s.5xlarge.12
+ |
+20
+ |
+240
+ |
+16/4.5
+ |
+280
+ |
+8
+ |
+4
+ |
+1 * H100 PCIe
+ |
+80
+ |
+KVM
+ |
+
+p5s.10xlarge.12
+ |
+40
+ |
+480
+ |
+24/9
+ |
+550
+ |
+16
+ |
+8
+ |
+2 * H100 PCIe
+ |
+160
+ |
+KVM
+ |
+
+p5s.20xlarge.12
+ |
+80
+ |
+960
+ |
+32/18
+ |
+750
+ |
+32
+ |
+8
+ |
+4 * H100 PCIe
+ |
+320
+ |
+KVM
+ |
+
+p5s.40xlarge.12
+ |
+160
+ |
+1920
+ |
+40/36
+ |
+850
+ |
+32
+ |
+8
+ |
+8 * H100 PCIe
+ |
+640
+ |
+KVM
+ |
+
+
+
+
+ P5s ECS Features
+ - 1:12 ratio of vCPUs to memory
- CPU: 4th Generation Intel® Xeon® Scalable 8458P processors (2.7 GHz of base frequency and 3.8 GHz of turbo frequency)
- Each GPU provides 80 GiB of GPU memory and 3,026 TFLOPS INT8 compute.
- The GPU memory bandwidth can reach up to 2,000 Gbit/s.
+ Supported Common Software
+ P5s 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 P5s ECSs. The following commonly used software is supported:
+ - Common deep learning frameworks, such as TensorFlow, Spark, PyTorch, MXNet, and Caffe
- CUDA GPU rendering supported by RedShift for Autodesk 3ds Max and V-Ray for 3ds Max
- Agisoft PhotoScan
- MapD
- More than 2,000 GPU-accelerated applications such as Amber, NAMD, and VASP
+ Notes
+ - P5s ECSs support automatic recovery when the hosts accommodating such ECSs become faulty.
- After a P5s ECS is stopped, basic resources (including vCPUs, memory, image, and encoding cards) are not billed, but its system disk is billed based on the disk capacity. If other products, such as EVS disks, EIP, and bandwidth are associated with the ECS, these products are billed separately.
- Specifications of P5s ECSs can only be changed to other specifications of the same instance type.
+ Computing-accelerated P3Overview
P3 ECSs use NVIDIA A100 GPUs and provide flexibility and ultra-high-performance computing. P3 ECSs have strengths in AI-based deep learning, scientific computing, Computational Fluid Dynamics (CFD), computing finance, seismic analysis, molecular modeling, and genomics. Theoretically, the FP32 is 19.5 TFLOPS and the TF32 tensor core is 156 TFLOPS | 312 TFLOPS (sparsity enabled).
Specifications
- Table 4 P3 ECS specificationsFlavor
+Table 5 P3 ECS specificationsFlavor
|
-vCPUs
+ | vCPUs
|
-Memory
+ | Memory
(GiB)
|
-Max./Assured Bandwidth (Gbit/s)
+ | Max./Assured Bandwidth (Gbit/s)
|
-Max. PPS
+ | Max. PPS
(10,000)
|
-Max. NIC Queues
+ | Max. NIC Queues
|
-Max. NICs
+ | Max. NICs
|
-GPUs
+ | GPUs
|
-GPU Memory
+ | GPU Memory
(GiB)
|
-Virtualization
+ | Virtualization
|
|---|
-p3.2xlarge.8
+ | p3.2xlarge.8
|
-8
+ | 8
|
-64
+ | 64
|
-10/4
+ | 10/4
|
-100
+ | 100
|
-4
+ | 4
|
-4
+ | 4
|
-1 × NVIDIA A100 80GB
+ | 1 × NVIDIA A100 80GB
|
-80
+ | 80
|
-KVM
+ | KVM
|
-p3.4xlarge.8
+ | p3.4xlarge.8
|
-16
+ | 16
|
-128
+ | 128
|
-15/8
+ | 15/8
|
-200
+ | 200
|
-8
+ | 8
|
-8
+ | 8
|
-2 × NVIDIA A100 80GB
+ | 2 × NVIDIA A100 80GB
|
-160
+ | 160
|
-KVM
+ | KVM
|
-p3.8xlarge.8
+ | p3.8xlarge.8
|
-32
+ | 32
|
-256
+ | 256
|
-25/15
+ | 25/15
|
-350
+ | 350
|
-16
+ | 16
|
-8
+ | 8
|
-4 × NVIDIA A100 80GB
+ | 4 × NVIDIA A100 80GB
|
-320
+ | 320
|
-KVM
+ | KVM
|
-p3.16xlarge.8
+ | p3.16xlarge.8
|
-64
+ | 64
|
-512
+ | 512
|
-36/30
+ | 36/30
|
-700
+ | 700
|
-32
+ | 32
|
-8
+ | 8
|
-8 × NVIDIA A100 80GB
+ | 8 × NVIDIA A100 80GB
|
-640
+ | 640
|
-KVM
+ | KVM
|
@@ -389,9 +517,9 @@
Supported Common Software
P3 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 P3 ECSs. P3 ECSs support the following commonly used software:
-- Common deep learning frameworks, such as TensorFlow, Spark, PyTorch, MXNet, and Caffee
- CUDA GPU rendering supported by RedShift for Autodesk 3dsMax and V-Ray for 3ds Max
- Agisoft PhotoScan
- MapD
- More than 2,000 GPU-accelerated applications such as Amber, NAMD, and VASP
-Notes
-- After a P3 ECS is stopped, basic resources (including vCPUs, memory, image, and GPUs) are not billed, but its system disk is billed based on the disk capacity. If other products, such as EVS disks, EIP, and bandwidth are associated with the ECS, these products are billed separately.
Resources will be released after a P3 ECS is stopped. If resources are insufficient at the next start, the start may fail. If you want to use such an ECS for a long period of time, do not stop the ECS.
+ - Common deep learning frameworks, such as TensorFlow, Spark, PyTorch, MXNet, and Caffe
- CUDA GPU rendering supported by RedShift for Autodesk 3ds Max and V-Ray for 3ds Max
- Agisoft PhotoScan
- MapD
- More than 2,000 GPU-accelerated applications such as Amber, NAMD, and VASP
+ Notes
+ - After a P3 ECS is stopped, basic resources (including vCPUs, memory, image, and GPUs) are not billed, but its system disk is billed based on the disk capacity. If other products, such as EVS disks, EIP, and bandwidth are associated with the ECS, these products are billed separately.
Resources will be released after a P3 ECS is stopped. If resources are insufficient at the next start, the start may fail. If you want to use such an ECS for a long period of time, do not stop the ECS.
- If a P3 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 Manually Installing a Tesla Driver on a GPU-accelerated ECS.
- GPU-accelerated ECSs differ greatly in general-purpose and heterogeneous computing power. Their specifications can only be changed to other specifications of the same instance type.
- GPU-accelerated ECSs do not support live migration.
@@ -399,129 +527,129 @@
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 5 P2s ECS specificationsFlavor
+Table 6 P2s ECS specificationsFlavor
|
-vCPUs
+ | vCPUs
|
-Memory
+ | Memory
(GiB)
|
-Max./Assured Bandwidth (Gbit/s)
+ | Max./Assured Bandwidth (Gbit/s)
|
-Max. PPS (10,000)
+ | Max. PPS (10,000)
|
-Max. NIC Queues
+ | Max. NIC Queues
|
-Max. NICs
+ | Max. NICs
|
-GPUs
+ | GPUs
|
-GPU Connection
+ | GPU Connection
|
-GPU Memory (GiB)
+ | GPU Memory (GiB)
|
-Virtualization
+ | Virtualization
|
-Hardware
+ | Hardware
|
|---|
-p2s.2xlarge.8
+ | p2s.2xlarge.8
|
-8
+ | 8
|
-64
+ | 64
|
-10/4
+ | 10/4
|
-50
+ | 50
|
-4
+ | 4
|
-4
+ | 4
|
-1 × V100
+ | 1 × V100
|
-PCIe Gen3
+ | PCIe Gen3
|
-1 × 32 GiB
+ | 1 × 32 GiB
|
-KVM
+ | KVM
|
-CPU: 2nd Generation Intel® Xeon® Scalable Processor 6278
+ | CPU: 2nd Generation Intel® Xeon® Scalable Processor 6278
|
-p2s.4xlarge.8
+ | p2s.4xlarge.8
|
-16
+ | 16
|
-128
+ | 128
|
-15/8
+ | 15/8
|
-100
+ | 100
|
-8
+ | 8
|
-8
+ | 8
|
-2 × V100
+ | 2 × V100
|
-PCIe Gen3
+ | PCIe Gen3
|
-2 × 32 GiB
+ | 2 × 32 GiB
|
-KVM
+ | KVM
|
-p2s.8xlarge.8
+ | p2s.8xlarge.8
|
-32
+ | 32
|
-256
+ | 256
|
-25/15
+ | 25/15
|
-200
+ | 200
|
-16
+ | 16
|
-8
+ | 8
|
-4 × V100
+ | 4 × V100
|
-PCIe Gen3
+ | PCIe Gen3
|
-4 × 32 GiB
+ | 4 × 32 GiB
|
-KVM
+ | KVM
|
-p2s.16xlarge.8
+ | p2s.16xlarge.8
|
-64
+ | 64
|
-512
+ | 512
|
-30/30
+ | 30/30
|
-400
+ | 400
|
-32
+ | 32
|
-8
+ | 8
|
-8 × V100
+ | 8 × V100
|
-PCIe Gen3
+ | PCIe Gen3
|
-8 × 32 GiB
+ | 8 × 32 GiB
|
-KVM
+ | KVM
|
@@ -533,137 +661,137 @@
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
+ 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 3ds Max and V-Ray for 3ds Max
- Agisoft PhotoScan
- MapD
- Notes- After a P2s ECS is stopped, basic resources (including vCPUs, memory, image, and GPUs) are not billed, but its system disk is billed based on the disk capacity. If other products, such as EVS disks, EIP, and bandwidth are associated with the ECS, these products are billed separately.
Resources will be released after a P2s ECS is stopped. If resources are insufficient at the next start, the start may fail. If you want to use such an ECS for a long period of time, do not stop the ECS.
+ Notes- After a P2s ECS is stopped, basic resources (including vCPUs, memory, image, and GPUs) are not billed, but its system disk is billed based on the disk capacity. If other products, such as EVS disks, EIP, and bandwidth are associated with the ECS, these products are billed separately.
Resources will be released after a P2s ECS is stopped. If resources are insufficient at the next start, the start may fail. If you want to use such an ECS for a long period of time, do not stop the ECS.
- By default, P2s ECSs created using a Windows 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 Manually Installing a Tesla Driver on a GPU-accelerated ECS.
- GPU-accelerated ECSs differ greatly in general-purpose and heterogeneous computing power. Their specifications can only be changed to other specifications of the same instance type.
- GPU-accelerated ECSs do not support live migration.
- Computing-accelerated P2vOverview
+ Computing-accelerated P2vOverview
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 6 P2v ECS specificationsFlavor
+Table 7 P2v ECS specificationsFlavor
|
-vCPUs
+ | vCPUs
|
-Memory
+ | Memory
(GiB)
|
-Max./Assured Bandwidth (Gbit/s)
+ | Max./Assured Bandwidth (Gbit/s)
|
-Max. PPS (10,000)
+ | Max. PPS (10,000)
|
-Max. NIC Queues
+ | Max. NIC Queues
|
-Max. NICs
+ | Max. NICs
|
-GPUs
+ | GPUs
|
-GPU Connection
+ | GPU Connection
|
-GPU Memory
+ | GPU Memory
(GiB)
|
-Virtualization
+ | Virtualization
|
-Hardware
+ | Hardware
|
|---|
-p2v.2xlarge.8
+ | p2v.2xlarge.8
|
-8
+ | 8
|
-64
+ | 64
|
-10/4
+ | 10/4
|
-50
+ | 50
|
-4
+ | 4
|
-4
+ | 4
|
-1 × V100
+ | 1 × V100
|
-N/A
+ | N/A
|
-1 × 16 GiB
+ | 1 × 16 GiB
|
-KVM
+ | KVM
|
-CPU: Intel® Xeon® Skylake-SP Gold 6151 v5
+ | CPU: Intel® Xeon® Skylake-SP Gold 6151 v5
|
-p2v.4xlarge.8
+ | p2v.4xlarge.8
|
-16
+ | 16
|
-128
+ | 128
|
-15/8
+ | 15/8
|
-100
+ | 100
|
-8
+ | 8
|
-8
+ | 8
|
-2 × V100
+ | 2 × V100
|
-NVLink
+ | NVLink
|
-2 × 16 GiB
+ | 2 × 16 GiB
|
-KVM
+ | KVM
|
-p2v.8xlarge.8
+ | p2v.8xlarge.8
|
-32
+ | 32
|
-256
+ | 256
|
-25/15
+ | 25/15
|
-200
+ | 200
|
-16
+ | 16
|
-8
+ | 8
|
-4 × V100
+ | 4 × V100
|
-NVLink
+ | NVLink
|
-4 × 16 GiB
+ | 4 × 16 GiB
|
-KVM
+ | KVM
|
-p2v.16xlarge.8
+ | p2v.16xlarge.8
|
-64
+ | 64
|
-512
+ | 512
|
-30/30
+ | 30/30
|
-400
+ | 400
|
-32
+ | 32
|
-8
+ | 8
|
-8 × V100
+ | 8 × V100
|
-NVLink
+ | NVLink
|
-8 × 16 GiB
+ | 8 × 16 GiB
|
-KVM
+ | KVM
|
@@ -676,9 +804,9 @@
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
+ 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 3ds Max and V-Ray for 3ds Max
- Agisoft PhotoScan
- MapD
- Notes- After a P2v ECS is stopped, basic resources (including vCPUs, memory, image, and GPUs) are not billed, but its system disk is billed based on the disk capacity. If other products, such as EVS disks, EIP, and bandwidth are associated with the ECS, these products are billed separately.
Resources will be released after a P2v ECS is stopped. If resources are insufficient at the next start, the start may fail. If you want to use such an ECS for a long period of time, do not stop the ECS.
+ Notes- After a P2v ECS is stopped, basic resources (including vCPUs, memory, image, and GPUs) are not billed, but its system disk is billed based on the disk capacity. If other products, such as EVS disks, EIP, and bandwidth are associated with the ECS, these products are billed separately.
Resources will be released after a P2v ECS is stopped. If resources are insufficient at the next start, the start may fail. If you want to use such an ECS for a long period of time, do not stop the ECS.
- By default, P2v ECSs created using a Windows public image have the Tesla driver installed.
- By default, P2v ECSs created using a Linux public image do not have a Tesla driver installed. After the ECS is created, install a driver on it for computing acceleration. For details, see Manually Installing a Tesla Driver on a GPU-accelerated ECS.
- 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 Manually Installing a Tesla Driver on a GPU-accelerated ECS.
- GPU-accelerated ECSs differ greatly in general-purpose and heterogeneous computing power. Their specifications can only be changed to other specifications of the same instance type.
- GPU-accelerated ECSs do not support live migration.
@@ -687,151 +815,151 @@
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 7 Pi2 ECS specificationsFlavor
+Table 8 Pi2 ECS specificationsFlavor
|
-vCPUs
+ | vCPUs
|
-Memory
+ | Memory
(GiB)
|
-Max./Assured Bandwidth
+ | Max./Assured Bandwidth
(Gbit/s)
|
-Max. PPS
+ | Max. PPS
(10,000)
|
-Max. NIC Queues
+ | Max. NIC Queues
|
-Max. NICs
+ | Max. NICs
|
-GPUs
+ | GPUs
|
-GPU Memory
+ | GPU Memory
(GiB)
|
-Local Disks
+ | Local Disks
|
-Virtualization
+ | Virtualization
|
-Hardware
+ | Hardware
|
|---|
-pi2.2xlarge.4
+ | pi2.2xlarge.4
|
-8
+ | 8
|
-32
+ | 32
|
-10/4
+ | 10/4
|
-50
+ | 50
|
-4
+ | 4
|
-4
+ | 4
|
-1 × T4
+ | 1 × T4
|
-1 × 16 GiB
+ | 1 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
-CPU: Intel® Xeon® Skylake 6151 3.0 GHz or Intel® Xeon® Cascade Lake 6278 2.6 GHz
+ | CPU: Intel® Xeon® Skylake 6151 3.0 GHz or Intel® Xeon® Cascade Lake 6278 2.6 GHz
|
-pi2.3xlarge.4
+ | pi2.3xlarge.4
|
-12
+ | 12
|
-48
+ | 48
|
-12/6
+ | 12/6
|
-80
+ | 80
|
-6
+ | 6
|
-6
+ | 6
|
-1 × T4
+ | 1 × T4
|
-1 × 16 GiB
+ | 1 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
-pi2.4xlarge.4
+ | pi2.4xlarge.4
|
-16
+ | 16
|
-64
+ | 64
|
-15/8
+ | 15/8
|
-100
+ | 100
|
-8
+ | 8
|
-8
+ | 8
|
-2 × T4
+ | 2 × T4
|
-2 × 16 GiB
+ | 2 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
-pi2.8xlarge.4
+ | pi2.8xlarge.4
|
-32
+ | 32
|
-128
+ | 128
|
-25/15
+ | 25/15
|
-200
+ | 200
|
-16
+ | 16
|
-8
+ | 8
|
-4 × T4
+ | 4 × T4
|
-4 × 16 GiB
+ | 4 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
-pi2.16xlarge.4
+ | pi2.16xlarge.4
|
-64
+ | 64
|
-256
+ | 256
|
-30/30
+ | 30/30
|
-400
+ | 400
|
-32
+ | 32
|
-8
+ | 8
|
-8 × T4
+ | 8 × T4
|
-8 × 16 GiB
+ | 8 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
diff --git a/docs/ecs/umn/en-us_topic_0107412162.html b/docs/ecs/umn/en-us_topic_0107412162.html
index e87cf9a7..37c453f7 100644
--- a/docs/ecs/umn/en-us_topic_0107412162.html
+++ b/docs/ecs/umn/en-us_topic_0107412162.html
@@ -1,6 +1,6 @@
-What Should I Do If Error "command ´gcc´ failed with exit status 1" Occurs During PIP-based Software Installation
+What Should I Do If Error "command ´gcc´ failed with exit status 1" Occurs During PIP-based Software Installation?
SymptomWhen installing the Python library software, you need to configure the PIP source. Take the official image source as an example: [root@test home]# cat /root/.pip/pip.conf
[global]
index-url = https://pypi.python.org/simple
diff --git a/docs/ecs/umn/en-us_topic_0116262067.html b/docs/ecs/umn/en-us_topic_0116262067.html
index 0f302e6e..a29a276b 100644
--- a/docs/ecs/umn/en-us_topic_0116262067.html
+++ b/docs/ecs/umn/en-us_topic_0116262067.html
@@ -1,16 +1,16 @@
-Key Operations Supported by CTS
+ECS Operations Supported by CTS
ScenariosCloud Trace Service (CTS) records user operations performed on ECSs and related resources for further query, auditing, and backtracking.
- PrerequisitesCTS has been provisioned.
+ PrerequisitesCTS has been enabled.
Key ECS Operations Recorded by CTS
Table 1 ECS operations recorded by CTSOperation
|
Resource Type
|
-Event Name
+ | Trace
|
|---|
diff --git a/docs/ecs/umn/en-us_topic_0116266206.html b/docs/ecs/umn/en-us_topic_0116266206.html
index 48c7264c..a09aa4f5 100644
--- a/docs/ecs/umn/en-us_topic_0116266206.html
+++ b/docs/ecs/umn/en-us_topic_0116266206.html
@@ -5,7 +5,7 @@
Constraints- Only stopped ECSs can be migrated.
- To ensure that the migration is successful, there must be an available DeH.
- CBR and CSBS backups are not affected by migrations.
- ECS IDs remain unchanged after a migration.
Procedure- Log in to the management console.
- Click
 in the upper left corner and select your region and project. - Under Computing, click Elastic Cloud Server.
- In the ECS list, locate the target ECS and choose More In the Operation column of the ECS to be migrated, choose More > Migrate ECS in the Operation column.
- In the displayed dialog box, select the target DeH.
If no DeHs are available, create a DeH first. For details, see Allocating DeHs in the Dedicated Host User Guide.
+ Procedure- Log in to the management console.
- Click in the upper left corner and select your region and project.
- Under Computing, click Elastic Cloud Server.
- In the ECS list, locate the target ECS to be migrated, choose More > Migrate ECS in the Operation column.
- In the displayed dialog box, select the target DeH.
If no DeHs are available, create a DeH first. For details, see Allocating DeHs in the Dedicated Host User Guide.
- Click OK.
diff --git a/docs/ecs/umn/en-us_topic_0140323150.html b/docs/ecs/umn/en-us_topic_0140323150.html
index 4fb0444c..6d655508 100644
--- a/docs/ecs/umn/en-us_topic_0140323150.html
+++ b/docs/ecs/umn/en-us_topic_0140323150.html
@@ -89,14 +89,14 @@
An exception has occurred on the ECS. Contact technical support for assistance.
|
- Reinstalling OS
+ | Reinstalling
|
Intermediate
|
The ECS has received a request to reinstall the OS and has begun the reinstallation.
|
- Reinstalling OS failed
+ | Reinstalling failed
|
Stable
|
@@ -110,7 +110,7 @@
The ECS received a request to change the OS and has begun implementing the changes.
|
- OS change failed
+ | Failed to change the OS
|
Stable
|
@@ -124,7 +124,7 @@
The ECS is being forcibly restarted.
|
- Rolling back resizing
+ | Reverting resizing
|
Intermediate
|
diff --git a/docs/ecs/umn/en-us_topic_0140323152.html b/docs/ecs/umn/en-us_topic_0140323152.html
index 049d2668..1592bc15 100644
--- a/docs/ecs/umn/en-us_topic_0140323152.html
+++ b/docs/ecs/umn/en-us_topic_0140323152.html
@@ -93,10 +93,10 @@
-  If the source is set to 0.0.0.0/0, any IP address can be used to remotely log in to the ECS. To ensure security, set the source to a specific IP address based on service requirements. For details about the configuration example, see Table 4.
+ If the source is set to 0.0.0.0/0, all external IP addresses are allowed to remotely log in to the ECS. To ensure network security and prevent service interruptions caused by network intrusions, set the source to a known IP address. For details, see Table 4.
- Table 4 Remotely logging in to an ECS using a specified IP addressECS Type
+Table 4 Remotely logging in to an ECS using a known IP addressECS Type
|
Direction
|
@@ -137,7 +137,7 @@
Remotely Connecting to an ECS from a Local Server to Upload or Download FTP FilesBy default, a security group denies all external requests. If you need to remotely connect to an ECS from a local server to upload or download files, you need to enable FTP ports 20 and 21.
- Table 5 Remotely connecting to an ECS from a local server to upload or download filesDirection
+Table 5 Remotely connecting to an ECS from any server to upload or download filesDirection
|
Type
|
@@ -159,12 +159,35 @@
|---|
You must first install the FTP server program on the ECSs and check whether ports 20 and 21 are working properly.
+ - If the source is set to 0.0.0.0/0, all external IP addresses are allowed to remotely log in to the ECS to upload or download files. To ensure network security and prevent service interruptions caused by network intrusions, set the source to a known IP address. For details, see Table 6.
- You must first install the FTP server program on the ECSs and check whether ports 20 and 21 are working properly.
+
+ Table 6 Remotely connecting to an ECS from a known server to upload or download filesDirection
+ |
+Type
+ |
+Protocol & Port
+ |
+Source
+ |
+
|---|
+
+Inbound
+ |
+IPv4
+ |
+TCP: 20-21
+ |
+IP address: 192.168.0.0/24
+ |
+
+
+
+ Setting Up a Website on an ECS to Provide Services ExternallyA security group denies all external requests by default. If you have set up a website on an ECS that can be accessed externally, you need to add an inbound rule to the ECS security group to allow access over specific ports, such as HTTP (80) and HTTPS (443).
- Table 6 Setting up a website on an ECS to provide services externallyDirection
+Table 7 Setting up a website on an ECS to provide services externallyDirection
|
Type
|
@@ -198,7 +221,7 @@
Using ping Command to Verify Network ConnectivityPing works by sending an Internet Control Message Protocol (ICMP) Echo Request. To ping an ECS from your PC to verify the network connectivity, you need to add an inbound rule to the security group of the ECS to allow ICMP traffic.
- Table 7 Using ping command to verify network connectivityDirection
+Table 8 Using ping command to verify network connectivityDirection
|
Type
|
@@ -232,7 +255,7 @@
Enabling Communications Between Instances in Different Security GroupsInstances in the same VPC but associated with different security groups cannot communicate with each other. If you want ECSs in security group sg-A to access MySQL databases in security group sg-B, you need to add an inbound rule to security group sg-B to allow access from ECSs in security group sg-A.
- Table 8 Enabling communications between instances in different security groupsDirection
+Table 9 Enabling communications between instances in different security groupsDirection
|
Type
|
@@ -258,7 +281,7 @@
Allowing External Instances to Access the Database Deployed on an ECSA security group denies all external requests by default. If you have deployed a database on an ECS and want the database to be accessed from external instances on a private network, you need to add an inbound rule to the security group of the ECS to allow access over corresponding ports. Here are some common ports for databases: - MySQL: port 3306
- Oracle: port 1521
- MS SQL: port 1433
- PostgreSQL: port 5432
- Redis: port 6379
- Table 9 Allowing external instances to access the database deployed on an ECSDirection
+Table 10 Allowing external instances to access the database deployed on an ECSDirection
|
Type
|
@@ -320,9 +343,9 @@
In this example, the source is for reference only. Set the source address based on your requirements.
-Allowing ECSs to Access Specific External WebsitesBy default, a security group allows all outbound traffic. Table 11 lists the default rules. If you want to allow ECSs to access specific websites, configure the security group as follows:
+ Allowing ECSs to Access Specific External WebsitesBy default, a security group allows all outbound traffic. Table 12 lists the default rules. If you want to allow ECSs to access specific websites, configure the security group as follows:
- Add outbound rules to allow traffic over specific ports to specific IP addresses.
-
Table 10 Allowing ECSs to access specific external websitesDirection
+Table 11 Allowing ECSs to access specific external websitesDirection
|
Type
|
@@ -360,7 +383,7 @@
|---|
- Delete the original outbound rules that allow all traffic.
-
Table 11 Default outbound rules in a security groupDirection
+Table 12 Default outbound rules in a security groupDirection
|
Type
|
diff --git a/docs/ecs/umn/en-us_topic_0140963099.html b/docs/ecs/umn/en-us_topic_0140963099.html
index ac6c4ecc..45966ee0 100644
--- a/docs/ecs/umn/en-us_topic_0140963099.html
+++ b/docs/ecs/umn/en-us_topic_0140963099.html
@@ -109,7 +109,7 @@
- Enable IPv6 for the ECS.
- Run the following command to check whether IPv6 is enabled for the kernel:
sysctl -a | grep ipv6
- If a command output is displayed, IPv6 is enabled.
- If no information is displayed, IPv6 is disabled. Go to 2.b to load the IPv6 module.
- Run the following command to load the IPv6 module:
modprobe ipv6
- - Add the following content to the /etc/sysctl.conf file:
net.ipv6.conf.all.disable_ipv6=0
+ - Add the following content to the /etc/sysctl.conf file:
net.ipv6.conf.all.disable_ipv6=0
- Save the configuration and exit. Then, run the following command to load the configuration:
sysctl -p
- Enable dynamic IPv6 address assignment for the ECS.
- Ubuntu 18.04/20.04
- Run the following command to access /etc/netplan/:
cd /etc/netplan
@@ -195,15 +195,15 @@ DHCPV6C=yes
- CentOS 6.x:
- Run the following command to edit the dhclient.conf file:
vi /etc/dhcp/dhclient.conf
- Press i to enter editing mode and add the timeout attribute to the file.
timeout 30;
- Enter :wq to save the settings and exit.
- - Debian 7.5:
- Run the following command to edit the networking file:
vi /etc/init.d/networking
+ - Debian 7.5:
- Run the following command to edit the networking file:
vi /etc/init.d/networking
- Press i to enter editing mode and add the timeout attribute.
Figure 20 Modification 1
Figure 21 Modification 2
- - Debian 8.2.0/8.8.0
- Run the following command to edit the network-pre.conf file:
vi /lib/systemd/system/networking.service.d/network-pre.conf
+ - Debian 8.2.0/8.8.0
- Run the following command to edit the network-pre.conf file:
vi /lib/systemd/system/networking.service.d/network-pre.conf
- Press i to enter editing mode and add the timeout attribute to the file.
[Service]
TimeoutStartSec=30
- - Debian 9.0
- Run the following command to edit the networking.service file:
vi /etc/system/system/network-online.target.wants/networking.service
+ - Debian 9.0
- Run the following command to edit the networking.service file:
vi /etc/system/system/network-online.target.wants/networking.service
- Press i to enter editing mode and change TimeoutStartSec=5min to TimeoutStartSec=30.
diff --git a/docs/ecs/umn/en-us_topic_0144542149.html b/docs/ecs/umn/en-us_topic_0144542149.html
index cff5476d..eee0a5cf 100644
--- a/docs/ecs/umn/en-us_topic_0144542149.html
+++ b/docs/ecs/umn/en-us_topic_0144542149.html
@@ -5,6 +5,8 @@
Each EIP can be used by only one cloud resource at a time.
Figure 1 Accessing the Internet using an EIP
+
diff --git a/docs/ecs/umn/en-us_topic_0163572589.html b/docs/ecs/umn/en-us_topic_0163572589.html
index 7c5e0f68..452dc29c 100644
--- a/docs/ecs/umn/en-us_topic_0163572589.html
+++ b/docs/ecs/umn/en-us_topic_0163572589.html
@@ -6,14 +6,15 @@
Basic Settings- Select an AZ.
An AZ is a physical location that uses independent power supply and networks. AZs in the same region can communicate with each other over an intranet.
- To enhance application availability, create ECSs in different AZs.
- To shorten network latency, create ECSs in the same AZ.
- - During the creation process, you can select a random AZ. The system will use a hash algorithm to select an AZ as the default AZ based on your universally unique identifier (UUID).
- The available ECS types and flavors vary depending on AZs. To view all supported ECS types and flavors on the cloud service platform, set AZ to Random. Then, the system automatically allocates an AZ according to your selected ECS flavor.
For example, S3 ECSs are available only in AZ1; S2 ECSs are available in AZ2 and AZ3 and have been sold out in AZ1. If you set AZ to Random, you can view both S3 and S2 ECSs. If you create an S3 ECS, the system automatically allocates it to AZ1. If you create an S2 ECS, the system randomly allocates it to AZ2 or AZ3.
+ - During the creation process, you can select a random AZ. The system will use a hash algorithm to select an AZ as the default AZ based on your universally unique identifier (UUID).
If you batch create multiple ECSs and want to deploy them in different AZs, select Multi-AZ deployment.
+ - The available ECS types and flavors vary depending on AZs. To view all supported ECS types and flavors on the cloud service platform, set AZ to Random. Then, the system automatically allocates an AZ according to your selected ECS flavor.
For example, S3 ECSs are available only in AZ1; S2 ECSs are available in AZ2 and AZ3 and have been sold out in AZ1. If you set AZ to Random, you can view both S3 and S2 ECSs. If you create an S3 ECS, the system automatically allocates it to AZ1. If you create an S2 ECS, the system randomly allocates it to AZ2 or AZ3.
- Set DeH.
This configuration is optional. This parameter is available only when you click Create ECS on the Dedicated Host page. It is unavailable when you click Create ECS on the Elastic Cloud Server page.
DeH refers to physical server resources dedicated for a specified user. You can deploy ECSs on DeHs for better isolation, security, and performance of your ECSs. You can continue using your existing server software licenses of ECSs on DeHs to reduce costs. For more details, see Dedicated Host User Guide.
- Set Specifications.
The cloud platform provides various ECS types for different application scenarios. You can choose from existing ECS types and flavors in the list. Alternatively, you can enter a flavor or specify vCPUs and memory size to search for the flavor suited to your needs.
Latest generation shows the types and flavors of newly released ECSs, and All generations show the types and flavors of all ECSs provided by the cloud platform.
- - Before selecting an ECS type, learn about the introduction and notes on each type of ECSs. For details, see ECS Types.
- You can select Hide sold-out specifications to hide specifications that have been sold out.
- Local Disk: specifies the local storage of the physical server where the ECS is deployed. Only Hard Disk Driver (HDD) disks are supported. If the ECS of the selected type (such as Disk-intensive) uses local disks, the system automatically attaches the local disks to the ECS and displays the information of the local disks.
For example, if Local Disk is 3x1800 GiB (HDD), three HDDs are attached to the ECS and the capacity of each HDD is 1800 GiB.
+
- Select an image.
- Public image
A public image is a standard, widely used image. It contains an OS and preinstalled public applications and is available to all users. You can configure the runtime environment or software in the public image as needed.
@@ -35,15 +36,15 @@
- Bring your own license (BYOL)
Allows you to use your existing OS license. In such a case, you do not need to apply for a license again.
For more information about license types, see License Types.
- - Set System Disk and Data Disk if required.
- - Set EIP.
An EIP is a static public IP address bound to an ECS in a VPC. Using the EIP, the ECS can provide services externally.
+ - Set EIP.
An EIP is a static public IP address bound to an ECS in a VPC. Using the EIP, the ECS can provide services externally.
The following options are provided:
- Auto assign
The system automatically assigns an EIP for the ECS. The EIP provides a dedicated bandwidth that is configurable.
- Use existing
An existing EIP is assigned for the ECS. When using an existing EIP, you are not allowed to create ECSs in a batch.
diff --git a/docs/ecs/umn/en-us_topic_0170265910.html b/docs/ecs/umn/en-us_topic_0170265910.html
index 89be1115..69024178 100644
--- a/docs/ecs/umn/en-us_topic_0170265910.html
+++ b/docs/ecs/umn/en-us_topic_0170265910.html
@@ -1,6 +1,6 @@
-Permissions Management
+Using IAM to Grant Access to ECS
diff --git a/docs/ecs/umn/en-us_topic_0170265913.html b/docs/ecs/umn/en-us_topic_0170265913.html
index 7d34d488..e1d6e546 100644
--- a/docs/ecs/umn/en-us_topic_0170265913.html
+++ b/docs/ecs/umn/en-us_topic_0170265913.html
@@ -19,7 +19,7 @@
diff --git a/docs/ecs/umn/en-us_topic_0170265914.html b/docs/ecs/umn/en-us_topic_0170265914.html
index b323bbc8..e6091c27 100644
--- a/docs/ecs/umn/en-us_topic_0170265914.html
+++ b/docs/ecs/umn/en-us_topic_0170265914.html
@@ -67,7 +67,7 @@
diff --git a/docs/ecs/umn/en-us_topic_0177512565.html b/docs/ecs/umn/en-us_topic_0177512565.html
index 957624c1..896a5cf4 100644
--- a/docs/ecs/umn/en-us_topic_0177512565.html
+++ b/docs/ecs/umn/en-us_topic_0177512565.html
@@ -3413,519 +3413,629 @@
|---|
Table 17 P3 ECS specificationsFlavor
+Table 17 P5s ECS specificationsFlavor
|
-vCPUs
+ | vCPUs
|
-Memory
-(GiB)
+ | Memory (GiB)
|
-Max./Assured Bandwidth (Gbit/s)
+ | Max./Assured Bandwidth (Gbit/s)
|
-Max. PPS
-(10,000)
+ | Max. PPS (10,000)
|
-Max. NIC Queues
+ | Max. NIC Queues
|
-Max. NICs
+ | Max. NICs
|
-GPUs
+ | GPUs
|
-GPU Memory
-(GiB)
+ | GPU Memory (GiB)
|
-Virtualization
+ | Virtualization
|
|---|
-p3.2xlarge.8
+ | p5s.5xlarge.12
|
-8
+ | 20
|
-64
+ | 240
|
-10/4
+ | 16/4.5
|
-100
+ | 280
|
-4
+ | 8
|
-4
+ | 4
|
-1 × NVIDIA A100 80GB
+ | 1 * H100 PCIe
|
-80
+ | 80
|
-KVM
+ | KVM
|
-p3.4xlarge.8
+ | p5s.10xlarge.12
|
-16
+ | 40
|
-128
+ | 480
|
-15/8
+ | 24/9
|
-200
+ | 550
|
-8
+ | 16
|
-8
+ | 8
|
-2 × NVIDIA A100 80GB
+ | 2 * H100 PCIe
|
-160
+ | 160
|
-KVM
+ | KVM
|
-p3.8xlarge.8
+ | p5s.20xlarge.12
|
-32
+ | 80
|
-256
+ | 960
|
-25/15
+ | 32/18
|
-350
+ | 750
|
-16
+ | 32
|
-8
+ | 8
|
-4 × NVIDIA A100 80GB
+ | 4 * H100 PCIe
|
-320
+ | 320
|
-KVM
+ | KVM
|
-p3.16xlarge.8
+ | p5s.40xlarge.12
|
-64
+ | 160
|
-512
+ | 1920
|
-36/30
+ | 40/36
|
-700
+ | 850
|
-32
+ | 32
|
-8
+ | 8
|
-8 × NVIDIA A100 80GB
+ | 8 * H100 PCIe
|
-640
+ | 640
|
-KVM
+ | KVM
|
Table 18 P2s ECS specificationsFlavor
+Table 18 P3 ECS specificationsFlavor
|
-vCPUs
+ | vCPUs
|
-Memory
-(GiB)
+ | Memory
+(GiB)
|
-Max./Assured Bandwidth (Gbit/s)
+ | Max./Assured Bandwidth (Gbit/s)
|
-Max. PPS (10,000)
+ | Max. PPS
+(10,000)
|
-Max. NIC Queues
+ | Max. NIC Queues
|
-Max. NICs
+ | Max. NICs
|
-GPUs
+ | GPUs
|
-GPU Connection
+ | GPU Memory
+(GiB)
|
-GPU Memory (GiB)
- |
-Virtualization
- |
-Hardware
+ | Virtualization
|
|---|
-p2s.2xlarge.8
+ | p3.2xlarge.8
|
-8
+ | 8
|
-64
+ | 64
|
-10/4
+ | 10/4
|
-50
+ | 100
|
-4
+ | 4
|
-4
+ | 4
|
-1 × V100
+ | 1 × NVIDIA A100 80GB
|
-PCIe Gen3
+ | 80
|
-1 × 32 GiB
+ | KVM
|
-KVM
+ |
+p3.4xlarge.8
|
-CPU: 2nd Generation Intel® Xeon® Scalable Processor 6278
+ | 16
+ |
+128
+ |
+15/8
+ |
+200
+ |
+8
+ |
+8
+ |
+2 × NVIDIA A100 80GB
+ |
+160
+ |
+KVM
+ |
+
+p3.8xlarge.8
+ |
+32
+ |
+256
+ |
+25/15
+ |
+350
+ |
+16
+ |
+8
+ |
+4 × NVIDIA A100 80GB
+ |
+320
+ |
+KVM
+ |
+
+p3.16xlarge.8
+ |
+64
+ |
+512
+ |
+36/30
+ |
+700
+ |
+32
+ |
+8
+ |
+8 × NVIDIA A100 80GB
+ |
+640
+ |
+KVM
+ |
+
+
+
+ Table 19 P2s ECS specificationsFlavor
+ |
+vCPUs
+ |
+Memory
+(GiB)
+ |
+Max./Assured Bandwidth (Gbit/s)
+ |
+Max. PPS (10,000)
+ |
+Max. NIC Queues
+ |
+Max. NICs
+ |
+GPUs
+ |
+GPU Connection
+ |
+GPU Memory (GiB)
+ |
+Virtualization
+ |
+Hardware
+ |
+
|---|
+
+p2s.2xlarge.8
+ |
+8
+ |
+64
+ |
+10/4
+ |
+50
+ |
+4
+ |
+4
+ |
+1 × V100
+ |
+PCIe Gen3
+ |
+1 × 32 GiB
+ |
+KVM
+ |
+CPU: 2nd Generation Intel® Xeon® Scalable Processor 6278
|
-p2s.4xlarge.8
+ | p2s.4xlarge.8
|
-16
+ | 16
|
-128
+ | 128
|
-15/8
+ | 15/8
|
-100
+ | 100
|
-8
+ | 8
|
-8
+ | 8
|
-2 × V100
+ | 2 × V100
|
-PCIe Gen3
+ | PCIe Gen3
|
-2 × 32 GiB
+ | 2 × 32 GiB
|
-KVM
+ | KVM
|
-p2s.8xlarge.8
+ | p2s.8xlarge.8
|
-32
+ | 32
|
-256
+ | 256
|
-25/15
+ | 25/15
|
-200
+ | 200
|
-16
+ | 16
|
-8
+ | 8
|
-4 × V100
+ | 4 × V100
|
-PCIe Gen3
+ | PCIe Gen3
|
-4 × 32 GiB
+ | 4 × 32 GiB
|
-KVM
+ | KVM
|
-p2s.16xlarge.8
+ | p2s.16xlarge.8
|
-64
+ | 64
|
-512
+ | 512
|
-30/30
+ | 30/30
|
-400
+ | 400
|
-32
+ | 32
|
-8
+ | 8
|
-8 × V100
+ | 8 × V100
|
-PCIe Gen3
+ | PCIe Gen3
|
-8 × 32 GiB
+ | 8 × 32 GiB
|
-KVM
+ | KVM
|
Table 19 P2v ECS specificationsFlavor
+Table 20 P2v ECS specificationsFlavor
|
-vCPUs
+ | vCPUs
|
-Memory
+ | Memory
(GiB)
|
-Max./Assured Bandwidth (Gbit/s)
+ | Max./Assured Bandwidth (Gbit/s)
|
-Max. PPS (10,000)
+ | Max. PPS (10,000)
|
-Max. NIC Queues
+ | Max. NIC Queues
|
-Max. NICs
+ | Max. NICs
|
-GPUs
+ | GPUs
|
-GPU Connection
+ | GPU Connection
|
-GPU Memory
+ | GPU Memory
(GiB)
|
-Virtualization
+ | Virtualization
|
-Hardware
+ | Hardware
|
|---|
-p2v.2xlarge.8
+ | p2v.2xlarge.8
|
-8
+ | 8
|
-64
+ | 64
|
-10/4
+ | 10/4
|
-50
+ | 50
|
-4
+ | 4
|
-4
+ | 4
|
-1 × V100
+ | 1 × V100
|
-N/A
+ | N/A
|
-1 × 16 GiB
+ | 1 × 16 GiB
|
-KVM
+ | KVM
|
-CPU: Intel® Xeon® Skylake-SP Gold 6151 v5
+ | CPU: Intel® Xeon® Skylake-SP Gold 6151 v5
|
-p2v.4xlarge.8
+ | p2v.4xlarge.8
|
-16
+ | 16
|
-128
+ | 128
|
-15/8
+ | 15/8
|
-100
+ | 100
|
-8
+ | 8
|
-8
+ | 8
|
-2 × V100
+ | 2 × V100
|
-NVLink
+ | NVLink
|
-2 × 16 GiB
+ | 2 × 16 GiB
|
-KVM
+ | KVM
|
-p2v.8xlarge.8
+ | p2v.8xlarge.8
|
-32
+ | 32
|
-256
+ | 256
|
-25/15
+ | 25/15
|
-200
+ | 200
|
-16
+ | 16
|
-8
+ | 8
|
-4 × V100
+ | 4 × V100
|
-NVLink
+ | NVLink
|
-4 × 16 GiB
+ | 4 × 16 GiB
|
-KVM
+ | KVM
|
-p2v.16xlarge.8
+ | p2v.16xlarge.8
|
-64
+ | 64
|
-512
+ | 512
|
-30/30
+ | 30/30
|
-400
+ | 400
|
-32
+ | 32
|
-8
+ | 8
|
-8 × V100
+ | 8 × V100
|
-NVLink
+ | NVLink
|
-8 × 16 GiB
+ | 8 × 16 GiB
|
-KVM
+ | KVM
|
Table 20 Pi2 ECS specificationsFlavor
+Table 21 Pi2 ECS specificationsFlavor
|
-vCPUs
+ | vCPUs
|
-Memory
+ | Memory
(GiB)
|
-Max./Assured Bandwidth
+ | Max./Assured Bandwidth
(Gbit/s)
|
-Max. PPS
+ | Max. PPS
(10,000)
|
-Max. NIC Queues
+ | Max. NIC Queues
|
-Max. NICs
+ | Max. NICs
|
-GPUs
+ | GPUs
|
-GPU Memory
+ | GPU Memory
(GiB)
|
-Local Disks
+ | Local Disks
|
-Virtualization
+ | Virtualization
|
-Hardware
+ | Hardware
|
|---|
-pi2.2xlarge.4
+ | pi2.2xlarge.4
|
-8
+ | 8
|
-32
+ | 32
|
-10/4
+ | 10/4
|
-50
+ | 50
|
-4
+ | 4
|
-4
+ | 4
|
-1 × T4
+ | 1 × T4
|
-1 × 16 GiB
+ | 1 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
-CPU: Intel® Xeon® Skylake 6151 3.0 GHz or Intel® Xeon® Cascade Lake 6278 2.6 GHz
+ | CPU: Intel® Xeon® Skylake 6151 3.0 GHz or Intel® Xeon® Cascade Lake 6278 2.6 GHz
|
-pi2.3xlarge.4
+ | pi2.3xlarge.4
|
-12
+ | 12
|
-48
+ | 48
|
-12/6
+ | 12/6
|
-80
+ | 80
|
-6
+ | 6
|
-6
+ | 6
|
-1 × T4
+ | 1 × T4
|
-1 × 16 GiB
+ | 1 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
-pi2.4xlarge.4
+ | pi2.4xlarge.4
|
-16
+ | 16
|
-64
+ | 64
|
-15/8
+ | 15/8
|
-100
+ | 100
|
-8
+ | 8
|
-8
+ | 8
|
-2 × T4
+ | 2 × T4
|
-2 × 16 GiB
+ | 2 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
-pi2.8xlarge.4
+ | pi2.8xlarge.4
|
-32
+ | 32
|
-128
+ | 128
|
-25/15
+ | 25/15
|
-200
+ | 200
|
-16
+ | 16
|
-8
+ | 8
|
-4 × T4
+ | 4 × T4
|
-4 × 16 GiB
+ | 4 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
-pi2.16xlarge.4
+ | pi2.16xlarge.4
|
-64
+ | 64
|
-256
+ | 256
|
-30/30
+ | 30/30
|
-400
+ | 400
|
-32
+ | 32
|
-8
+ | 8
|
-8 × T4
+ | 8 × T4
|
-8 × 16 GiB
+ | 8 × 16 GiB
|
-N/A
+ | N/A
|
-KVM
+ | KVM
|
diff --git a/docs/ecs/umn/en-us_topic_0240831198.html b/docs/ecs/umn/en-us_topic_0240831198.html
index 7a025b10..e15a5703 100644
--- a/docs/ecs/umn/en-us_topic_0240831198.html
+++ b/docs/ecs/umn/en-us_topic_0240831198.html
@@ -10,7 +10,7 @@
UUID=b9a07b7b-9322-4e05-ab9b-14b8050bdc8a / ext4 defaults 0 1
/dev/vdb1 /data1 ext4 defaults 0 0
/dev/vdc1 /data2 ext4 defaults 0 0
-After /dev/vdb1 is detached and the ECS is restarted, /dev/vdc1 becomes /dev/vdb1 and is mounted to /data. In such a case, no disk is mounted to /data2.
+After /dev/vdb1 is detached and the ECS is restarted, /dev/vdc1 becomes /dev/vdb1 and is mounted to /data1. In such a case, no disk is mounted to /data2.
The change of drive letters can affect the running of applications. To solve this problem, you are advised to use the universally unique identifiers (UUIDs) to replace /dev/vdx because a UUID uniquely identifies a disk partition in the Linux OS.
Solution- Log in to the ECS.
- Run the following command to obtain the partition UUID:
blkid Disk partition
|
|---|
|
|---|
|
|---|
|
|---|
|
|---|
|
|---|
|---|
|
|---|
|---|
|
|---|
|---|
|
|---|
|---|
|
|---|
|
|---|
|---|
|
|---|
|
|---|
|
|---|
|
|---|
|
|---|
|
