Support Rocky Linux 9.3 OS, and allow users to create BCC and EBC nodes using Rocky Linux 9.3 to enhance user experience.
|
2024-11
|
The console supports creating secrets via forms, eliminating the need for base64 encoding and reducing usage thresholds.
Refresh the left navigation bar for online cluster management.
The node list supports filtering cordoned nodes to optimize the user experience.
Support RBAC authorization for IAM roles, grant cluster RBAC permissions to IAM roles, and enable users to access cluster resources through roles. For more information, please refer to Configure Preset RBAC Permission Policies.
Optimize the cluster security group to support the configuration of automatically creating default security groups and using existing security groups. For more information, please refer to CCE Security Group.
Launch a default node resource reservation policy to support automatic adjustment of system-reserved resources based on node specification size and improve node stability. For more information, please refer to CCE Node Resource Reservation Description.
|
|
2024-10
|
Add a new Cluster Inspection function to help you identify potential risks in container service clusters.
Add a new VPC-ENI cluster to support baremetal BBC node function and add baremetal BBC as nodes. In VPC-ENI mode, allocate container IP addresses as secondary IP addresses of the primary network interface card.
|
|
2024-09
|
Add node support to specify an Elastic High-performance Computing Cluster (EHC). Nodes in the same EHC cluster can communicate via RDMA networks with high bandwidth and low latency. For more information, please refer to Elastic High-Performance Cluster Overview.
Node groups support binding self-healing rules, enabling configuration of failure detection items and self-healing operations to automatically detect and execute self-healing operations when node faults occur. For more information, please refer to Node Group Failure Detection and Self-Healing.
Add the CCE Credential Controller component for password-free image pull to component management, eliminating the need to specify private image access credentials when deploying workloads.
Add a search function to component management to support quick component retrieval.
The CCE Ingress Nginx Controller component supports K8S v1.26 or higher.
|
|
2024-08
|
Release Kubernetes v1.30, and adhere to K8S consistency principles. For more information, please refer to CCE Release of Kubernetes v1.30 History. Low-version clusters can be upgraded to v1.30.
A backup center module is launched, providing an integrated solution for backup and recovery of containerized applications. For more information, please refer to Backup Overview.
Add a new resource object browser function to support management of Kubernetes resources (including custom resources) by API group and version No.
Add a new node draining operation to support active eviction of Pods on nodes and cordoning nodes and improve node operation and maintenance capabilities.
Add a virtual node to the node list to support management of BCI Pods scheduled on virtual nodes.
The CCE CronHPA Controller component supports Kubernetes 1.28 and above and timezone configuration.
|
|
2024-07
|
Prometheus monitor pre-configures cluster component monitor panels, providing basic metric displays for network, storage, monitor, log, image and other components.
The new version of the CCE Ingress Controller component in managed clusters will be hosted on the CCE side, with zero operation and maintenance for users.
Add core components (Kube APIServer, Kube Scheduler and Kube Controller Manager) of managed clusters to component management to support quick enabling of logstore and container monitor for core components.
CoreDNS components are connected to the log center, supporting one-click logstore and providing CoreDNS log persistence and analysis capabilities.
New nodes support specifying directories to mount local disks, enabling local models to use local disks for container data storage.
The restriction that nodes and containers cannot use the same private network subnet is lifted in the VPC-ENI container network mode.
|
|
2024-06
|
Release Kubernetes v1.28, and adhere to K8S consistency principles. For more information, please refer to CCE Release of Kubernetes v1.28 History.
Container runtime Containerd supports v1.7.13.
The CCE CSI CDS Plugin component supports configuring whether a dynamically mounted cloud disk server (CDS) enters the bucket trash when deleted. For more information, please refer to CCE CSI CDS Plugin Version History
The API Server of newly created managed clusters is adjusted to bind users’ BLB and EIP resources, supporting cluster access control by configuring security group rules for BLB.
The event center provides an event overview dashboard, including total events, distribution, node anomalies, Pod OOMs, trends of important events and other dashboards as well as a list of top abnormal events.
Cluster audit provides visualization dashboards for audit logs, offering statistical analysis of key audit logs. For more information, please refer to Cluster Audit Dashboard.
|
|
2024-05
|
Support the construction of standard independent clusters by the managed edge computing nodes, and provide container deployment capabilities for edge services.
Launch the CCE CSI PFS L2 Plugin component to support mounting of parallel filesystem service (L2 Extreme PFS) to containers.
Node groups support Add Existing Nodes, allowing ungrouped nodes in the cluster and external servers to be added to node groups for unified management.
|
|
2024-04
|
The CCE CronHPA Controller component is launched, providing scheduled auto scaling capabilities for workloads to meet resource elasticity requirements of periodically changing services and reduce resource costs.
Cluster auditing supports delivery to Baidu AI Cloud Log Service (BLS) for persistent storage and query.
Add existing nodes to support formatting server data disks and specifying data directories for mounting, enabling data disks to store container and kubelet data.
|
|
2024-03
|
The console supports viewing Pods running on nodes, facilitating Pod management from a node perspective.
Component management launches the CoreDNS component, supporting manual upgrade of component versions.
|
|
2024-02
|
LB Service supports multi-protocol deployment on the same port to meet service needs, such as DNS.
Release Kubernetes v1.26, and adhere to K8S consistency principles. For more information, please refer to CCE Release of Kubernetes v1.26 History.
Add a replication operation to node groups to support the quick creation of new node groups based on existing node group configurations.
|
|
2023-11
|
The NodeLocalDNSCache component is launched to improve the stability and performance of cluster service discovery.
Clusters support container resource isolation via cGroup v2.
LB Service/Ingress supports specifying EIP internet connection types and public network bandwidth ranges via Annotation.
The CCE cluster-integrated monitor service (CProm) pre-configures Prometheus monitor planes, including control plane component monitor (APIServer, ETCD, Kube Scheduler and Kube Controller Manager), node monitor, workload monitor, network monitor (CoreDNS) and cluster Namespace monitor.
CCE clusters pre-configure cloud-native AI monitor planes, including resource statistics, node resources, workloads, scheduler component stability and GPUManager stability. These planes help understand GPU usage, workload performance, and scheduler component stability to quickly diagnose abnormalities, optimize GPU resource allocation and improve resource utilization.
|
|
2023-10
|
The CCE Qos Agent component is launched, providing rich single-machine isolation capabilities to improve cluster resource utilization and stability.
Log collection rules support Configure the Delivery Endpoint BES, transferring container logs to BES for storage and analysis.
The failure self-healing component CCE Node Remedier supports configuring self-healing policies for restarting elastic baremetal compute (EBC).
|
|
2023-09
|
CCE CSI BOS Plugin supports mounting Baidu AI Cloud Object Storage (BOS) in Containerd runtime environments.
The fault detection component CCE Node Problem Detector supports detection of hardware faults such as CPU and memory.
LB Service supports selecting and mounting backend target nodes based on labels.
|
|
2023-08
|
CCE clusters integrate resource labels and Tag-based Authentication, supporting cluster management and authorization via tags.
The storage component CCE CSI PFS Plugin releases v1.2.5, supporting multiple PFS instances. For more information, please refer to Use Parallel Filesystem Service (PFS).
Add a new Load-aware Scheduler Plugin to support scheduling based on actual node resource usage.
LB Service/Ingress supports direct modification of associated BLB and EIP configuration parameters via YAML. For specific configurations, please refer to LoadBalancer Service Annotation Description and BLB Ingress Annotation Description.
|
|
2023-07
|
IPv6 dual-stack networks are implemented based on the VPC-ENI container network mode, supporting deployment of IPv4 and IPv6 services.
The console supports creating elastic baremental compute (EBC) nodes using custom images directly.
|
|
2023-06
|
VPC-ENI container network mode supports elastic baremental compute (EBC).
Add the CCE Descheduler rescheduling component to the component management center to support Pod rescheduling based on real node utilization.
Add GPU and other hardware fault detection to the CCE Node Problem Detector fault detection component. For more information, please refer to Introduction to GPU Fault Detection.
When deploying workloads, the console supports directly selecting container images from the Cloud Container Registry (CCR) (enterprise and personal).
|
|
2023-05
|
Control plane component logs of managed clusters are connected to Baidu AI Cloud Log Service (BLS). For more information, please refer to View Cluster Control Plane Logs.
Add node support for custom Kubelet parameters. For more information, please refer to Custom Kubelet Parameters.
Support updating cluster access credential certificates.
Support enabling network policy when creating a cluster. For more information, please refer to Use Network Policy.
VPC route container network mode supports elastic baremental compute (EBC).
VPC-ENI container network mode supports Pod-exclusive elastic network interfaces, with allow list access available.
|
|
2023-04
|
Kubernetes v1.24 is released. For more information, please refer to CCE Release of Kubernetes v1.24 History.
Add the CCE Node Remedier failure self-healing component to the component management center.
The console supports setting namespace resource quotas and limits. For more information, please refer to Set Resource Quotas and Set Limit Ranges.
Support creating cluster nodes using the public image BaiduLinux 3.0.
LB Service/Ingress supports configuring EIP internet connection types.
|
|
2023-03
|
Clusters support configuring custom IAM permission policies. For more information, please refer to Configure Custom IAM Permission Policies.
Add the CCE Node Problem Detector failure detection component to the component management center.
LB Service (local mode) supports load balancer configuration. For more information, please refer to Set Dynamic Weights for Server Groups.
GPU node creation supports checking to install drivers.
|
|
2023-02
|
Add a new log center function to connect container logs to Baidu AI Cloud Log Service (BLS). For more information, please refer to Log Overview.
Clusters and node groups support creating prepaid nodes.
Support creating cluster nodes using the public image Rocky Linux 8.6.
LB Service supports extending protocols such as Http and Https. For more information, please refer to Extend Service Port Protocols.
Node groups support uniform distribution of multiple subnets and scaling strategies. For more information, please refer to Configure Scaling Policies.
Add a new component management function to support the management of cluster extension components. For more information, please refer to Component Overview.
|
|
2023-01
|
The console supports viewing container subnets.
[Experience optimization] The product page style is fully upgraded; you can experience it on the console.
|
|
2022-06
|
Node groups support Add Standby Models, which can specify different models (with the same specifications) in different availability zones.
|
|
2022-05
|
Support changing node group configuration.
Support BLB Ingress Annotation to enable richer load balancing capabilities.
|
|
2022-04
|
The Mixed Deployment Dashboard adds metrics such as mixed deployment benefit calculation and container utilization statistics.
Optimize offline workload creation and resource quota setting processes.
|
|
2021-08
|
Add Namespace Limit Range Description. Cluster administrators can set resource limits for each tenant via resource quota and limit range.
|
|
2021-07
|
Add Scale-up Scheme when the container network segment space is exhausted in VPC-CNI mode. In VPC-CNI mode, container IPs are allocated from subnets in the VPC. As the cluster scales, IPs in ENI subnets may be exhausted. By utilizing the dynamic configuration function provided by CNI, new ENI subnets can be added to the cluster.
|
|
2020-09
|
Adjust the menu position of console application management-related functions (workload, traffic ingress, storage configuration, monitoring logs, etc.) and move them to the single cluster management page. For details, please refer to CCE Console Upgrade
Serverless Kubernetes Clusters are open for public beta, simplifying cluster management and operation, and reducing deployment and maintenance costs
Cluster creation supports selecting different types of templates based on actual business scenarios
Support using Key Management Service (KMS) to encrypt Kubernetes Secret data stored in etcd
|
|
2020-08
|
Upgrade of cluster management function: (1) Support custom master multi-replica specifications and multi-availability zones; (2) Support creating or scaling clusters with existing master and worker instances, and provide custom script injection function; (3) Support specifying VPC subnets for master, LB service and ingress-associated Baidu load balance (BLB); (4) Kubelet is upgraded to dynamic configuration, and KubeProxy is deployed using containerization; (5) Support node cordoning, cluster OIDC authentication and selection of cluster NodeName types
Enhanced container network function: (1) Support custom Service ClusterIP and Service NodePort; (2) Support custom network interface card types (kubenet, ipvlan, veth and auto) in VPC network mode
Optimized interaction experience: (1) New cluster management OpenAPI is released; (2) The console cluster creation interface is upgraded
|
|
2020-07
|
CCE deployment supports custom K8S parameters
VPC-CNI container network mode supports fixed Pod IPs
Support migration of Cloud Container Engine (CCE) images to Cloud Container Registry (CCR)
|
|
2020-06
|
The CCE console supports WebShell
Support Set Periodic and Scheduled Scaling Tasks for Deployment and StatefulSet, and expand the application scenarios of auto scaling combined with workload automatic Horizontal Pod Autoscaler (HPA)
|
|
2020-05
|
Auto scaling of clusters supports configuring spot instance waiting time
Optimize console prompt explanation document and document link display
|
|
2020-04
|
Support Kubernetes v1.16.8
Container monitor alarm configuration is integrated with IAM, supporting sending alarms to specified users or user groups via email or SMS
Launch the event center function
Node configuration supports selecting BCC instance families
The CCE console supports configuring workload automatic Horizontal Pod Autoscaler
The CCE console supports creating and managing virtual nodes
|
|
2020-03
|
Support saving YAML files for creating workloads, services and storage configurations as custom templates for management, editing and deletion
Support deploying workloads using Docker images via forms
Support creating persistent volumes and persistent volume claims via forms
|
|
2020-02
|
Support creating clusters using Kunlun Baidu Cloud Compute (Baidu’s self-developed XPU chips)
Optimize cluster scale-up strategy and display the number of nodes in the scalability group
|
|
2020-01
|
Support auto scaling of cluster spot instances using scalability groups
|
|
2019-12
|
Container network supports the VPC-CNI mode, setting the container network and cluster node network to the same network address space
Container logs support pushing to Baidu AI Cloud Object Storage (BOS)
|
|
2019-11
|
Support auto scaling of clusters using scalability groups
|
|
2019-10
|
Support assigning Kubernetes RBAC permissions in K8S Cluster CCE to Baidu AI Cloud IAM Users
CCE supports parallel processing of cluster scaling requests
|
|
2019-09
|
Support adding Physical Servers (BBC) as K8S Cluster CCE nodes
|
|
2019-08
|
CCE supports configuring RBAC Permissions for Sub-Account with CCE read-only permissions to implement permission control over K8S Cluster CCE namespaces
|
|
2019-07
|
CCE allows users to use the open-source tool velero for backup and recovery of Kubernetes clusters on the cloud. It applies to scenarios such as cluster misoperations, cluster failures and cluster migrations
|
|
2019-06
|
-
Support K8S operation auditing
-
Optimize the cluster list and display the usage of newly added resources
|
|
2019-05
|
-
Support independent master deployment
-
Launch the alarm function
-
Support K8S version upgrade
|
|
2019-04
|
-
Support Helm template management and Helm application installation
-
Support moving existing nodes in and out
-
Support prepaid worker nodes
-
Support cluster event progress tracking
|
|
2019-03
|
Support using container image to build services. Baidu AI Cloud Image Service allows users to configure image building rules based on code repositories or Dockerfiles on the page. Image building can be triggered manually by users or automatically when code repositories change, connecting the intermediate link from user code changes to container deployment, so that users can deploy containerized services more efficiently and conveniently
|
|
2019-02
|
-
Support ingress traffic management
-
UI interface upgraded with new menus: Configure Storage, Monitor Logs and Traffic Ingress
-
Image registry supports automatic image building
|
|
2018-12
|
-
Support Baidu AI Cloud identity and access management
-
Support container log collection. Users can set management policies for container logs and service logs in Kubernetes on the CCE console
-
Support cluster resource and master component monitor
-
Users can mount BOS storage to containers for use via PV/PVC in Kubernetes
|
|
2018-11
|
Support users to quickly create and manage application BLB load balancing for backend services deployed in clusters on the CCE console
|
|
2018-10
|
-
Support selecting K8S version when creating a cluster
-
Support creating clusters using DCC
-
Launch Fabric and Xuper Chain solutions
-
Launch application management, and support console management of containers, services, deployments and other resources
|
|
2018-08
|
-
Launch Baidu AI Cloud in North China - Baoding region
-
Limit the minimum node configuration to 4 G and 8 GB
-
Upgrade K8S version to 1.11
-
Open node billing change function
-
Integrate Prometheus and alert for monitoring and alert
-
Mount multiple CDS disks in a single Pod
|
|
2018-05
|
-
Launch Baidu AI Cloud in Hong Kong region
-
Launch auto scaling function
-
Nodes support CentOS images
-
Introduce log and event mechanisms
-
Launch K8S dashboard
-
Support using Baidu AI Cloud CDS and CFS storage via PV
|
|
2018-03
|
-
Launch Baidu AI Cloud in Guangzhou and Suzhou regions
-
CCE Service high availability
-
Upgrade K8S version to 1.8
-
Support NAT subnets
-
Obtain CNCF "Kubernetes Conformance Certification"
|
|
2017-12
|
-
User cluster master nodes with 3-replica high availability
-
Open Baidu AI Cloud Image Registry
-
Support CDS disk pre-mounting
-
Support GPU nodes
|
|
2017-08
|
Officially release Baidu AI Cloud’s Cloud Container Engine (CCE)
|
