Kingsoft Cloud Serverless Kubernetes
Based on Kingsoft Cloud Container Instance (KCI), Kingsoft Cloud Serverless Kubernetes (KSK) provides a container service that is fully compatible with the Kubernetes ecosystem. You can directly deploy containerized applications and manage them in a Kubernetes-native way without the need to purchase or manage underlying nodes. This facilitates the deployment of Kubernetes applications.
Benefits
Ease of Use
You can quickly create fully managed serverless clusters without the need to purchase or manage underlying nodes. This enables you to focus more on your applications.
Kubernetes-native Support
KSK is fully compatible with Kubernetes-native cluster management, which enables you to seamlessly migrate Kubernetes applications to KSK.
Ecosystem Compatibility
KSK is deeply integrated with other Kingsoft Cloud products, such as Virtual Private Cloud (VPC), Server Load Balancing (SLB), and Elastic Block Storage (EBS), to provide solutions with integrated public cloud capabilities.
Secure Isolation
KSK provides the secure isolation capability of virtual machines based on the secure container technology implemented on KCI.
Features
Cluster Management
KSK provides full lifecycle management for clusters and Kubernetes applications. You can manage various resources such as deployment, Horizontal Pod Autoscaler (HPA), service, configuration, and storage.
Pay-As-You-Go
KSK enables you to select the CPU and memory specifications based on the workload. You are billed for the amount of time you use resources, in seconds. You do not need to pay for idle computing resources.
Container Scheduling
KSK integrates container instances with serverless clusters by using virtual nodes. It also supports custom workload scheduling rules.
Rolling Upgrade
KSK supports rolling upgrade and rollback to ensure continuous business during upgrade.
Scenarios
Online Business
Offline Computing
Real-time Audio and Video Processing
Online Business
Based on the monitoring metrics of clusters and the native HPA of Kubernetes, KSK supports efficient and elastic scaling of instances based on actual loads of applications. This reduces O&M costs and improves service efficiency.
Efficient scale-out
Compared with node scale-out, scale-out of container instances is more efficient to cope with traffic bursts of online business in a timely manner.
Simplified O&M
You can quickly create fully managed clusters without the need to manage underlying nodes. This helps simplify the O&M process.
Cost reduction
For online business with significant traffic fluctuations, the pay-as-you-go billing mode greatly reduces the cost.
Related Services
Offline Computing
For the offline batch processing of static data, KSK uses Job and CronJob resources to perform one-time or scheduled periodic tasks, such as offline reports and data analysis.
Kubernetes-native support
KSK supports the native management mode and pluggable capability of Kubernetes to meet the requirements for various task models in batch data processing.
Pay-as-you-go
For scheduled periodic tasks, you are billed only for the resources that are consumed by the workload. You do not need to pay for idle resources.
Related Services
Real-time Audio and Video Processing
KSK supports real-time audio and video processing tasks, such as encoding, decoding, and transcoding, in the live streaming acceleration service. KSK meets the requirements for stable and smooth playback and high concurrency in live streaming scenarios.
Fast scale-out
KSK scales out container instances in seconds, freeing you from considering the node scale-out cycle.
Ecosystem compatibility
KSK is compatible with the Kingsoft Cloud ecosystem and can be integrated with various other Kingsoft Cloud products related to network, storage, and CDN to provide comprehensive solutions.