Tencent Kubernetes Engine
- Release Notes and Announcements
- Release Notes
- Announcements
- Security Vulnerability Fix Description
- Release Notes
- Kubernetes Version Maintenance
- Runtime Version Maintenance Description
- Product Introduction
- Quick Start
- TKE General Cluster Guide
- Purchase a TKE General Cluster
- Permission Management
- Controlling TKE cluster-level permissions
- TKE Kubernetes Object-level Permission Control
- Cluster Management
- Images
- Worker node introduction
- Normal Node Management
- Native Node Management
- Purchasing Native Nodes
- Supernode management
- Purchasing a Super Node
- Registered Node Management
- Memory Compression Instructions
- GPU Share
- qGPU Online/Offline Hybrid Deployment
- Kubernetes Object Management
- Workload
- Configuration
- Auto Scaling
- Service Management
- Ingress Management
- CLB Type Ingress
- API Gateway Type Ingress
- Storage Management
- Use File to Store CFS
- Use Cloud Disk CBS
- Add-On Management
- CBS-CSI Description
- Network Management
- GlobalRouter Mode
- VPC-CNI Mode
- Static IP Address Mode Instructions
- Cilium-Overlay Mode
- OPS Center
- Audit Management
- Event Management
- Monitoring Management
- Log Management
- Using CRD to Configure Log Collection
- Backup Center
- TKE Serverless Cluster Guide
- Purchasing a TKE Serverless Cluster
- TKE Serverless Cluster Management
- Super Node Management
- Kubernetes Object Management
- OPS Center
- Log Collection
- Using a CRD to Configure Log Collection
- Audit Management
- Event Management
- TKE Registered Cluster Guide
- Registered Cluster Management
- Ops Guide
- TKE Insight
- TKE Scheduling
- Job Scheduling
- Native Node Dedicated Scheduler
- Cloud Native Service Guide
- Cloud Service for etcd
- Version Maintenance
- Cluster Troubleshooting
- Practical Tutorial
- Cluster
- Serverless Cluster
- Mastering Deep Learning in Serverless Cluster
- Scheduling
- Security
- Service Deployment
- Proper Use of Node Resources
- Network
- DNS
- Self-Built Nginx Ingress Practice Tutorial
- Logs
- Monitoring
- OPS
- DevOps
- Construction and Deployment of Jenkins Public Network Framework Appications based on TKE
- Auto Scaling
- KEDA
- Containerization
- Microservice
- Cost Management
- Hybrid Cloud
- Fault Handling
- Pod Status Exception and Handling
- Pod exception troubleshooter
- API Documentation
- Making API Requests
- Elastic Cluster APIs
- Resource Reserved Coupon APIs
- Cluster APIs
- Third-party Node APIs
- Network APIs
- Node APIs
- Node Pool APIs
- TKE Edge Cluster APIs
- Cloud Native Monitoring APIs
- Super Node APIs
- TKE API 2022-05-01
- Making API Requests
- Node Pool APIs
- FAQs
- TKE General Cluster
- TKE Serverless Cluster
- About OPS
- Service Agreement
- User Guide(Old)
DocumentationTencent Kubernetes EngineTKE General Cluster Guide Network Management VPC-CNI ModeLimits on the Number of Pods in VPC-CNI Mode
Limits on the Number of Pods in VPC-CNI Mode
Last updated: 2024-12-19 21:04:06
This document describes the default limits on the number of Pods in different VPC-CNI network modes.
Limits on the Number of Pods with Shared ENI
The number of Pods with shared ENI is limited by the number of ENIs that can be bound to a node and the number of IPs that can be bound on a single ENI. By default, the maximum number of Pod IPs on a single node with multiple ENIs = the maximum number of secondary ENIs that can be bound * the number of secondary IPs that can be bound on a single ENI, and the maximum number of Pod IPs on a single node with a single ENI = the number of secondary IPs that can be bound on a single ENI. See the table below for details.
CPU Cores | 1 | 2-6 | 8-10 | >=12 |
The maximum number of secondary ENIs that can be bound | 1 | 3 | 5 | 7 |
The number of secondary IPs that can be bound on a single ENI | 5 | 9 | 19 | 29 |
The maximum number of Pod IP addresses on a single node (with multiple ENIs) | 5 | 27 | 95 | 203 |
The maximum number of Pod IP addresses on a single node (with a single ENI) | 5 | 9 | 19 | 29 |
Note:
Multi-ENI component versions are supported (v3.3 or later in non-static IP address mode, and v3.4 or later in static IP address mode).
The number of ENIs that can be bound to each model and the number of IPs that can be bound on a single ENI is slightly different. For details, see Use Limits.
Limits on the Number of Pods with Exclusive ENIs
The number of Pods with exclusive ENIs is only limited by the number of ENIs that can be bound to the node. It only supports some models such as S5, SA2, IT5 and SA3. See the table below for details.
CPU Cores Model | 1 | 2 | 4 | >=8 | >=128 |
S5 | 4 | 9 | 19 | 39 | 23 |
SA2 | 4 | 9 | 19 | 39 | 23 |
IT5 | 4 | 9 | 19 | 39 | 23 |
SA3 | 4 | 9 | 15 | 15 | 15 |
Was this page helpful?
You can also Contact Sales or Submit a Ticket for help.
Yes
No