Introduction to Third-Party Frameworks
Integrate your CKS clusters with third-party tools and services
You can integrate CKS clusters with a variety of third-party frameworks to make running jobs and workloads more efficient and effective. Below are common frameworks:
| Framework | Description |
|---|---|
| Ray | Ray is an open-source framework for scaling AI and Python applications. As a distributed computing framework for parallel and distributed applications, Ray allows you to scale your applications across multiple Nodes in a CKS cluster, making it ideal for machine learning, data processing, and other compute-intensive tasks. For more information about Ray, refer to the Ray documentation. |
| SGLang | SGLang is a fast serving framework for large language models and vision language models. It makes your interaction with models faster and more controllable by co-designing the backend runtime and frontend language. For more information about SGLang, refer to the SGLang documentation. |
Orchestration tools for implementing frameworks
To run the frameworks, we use some of the following tools for orchestrating jobs on CKS:
-
Kueue: Kueue is a Kubernetes-native job queueing system that helps manage and schedule workloads in a CKS cluster. It allows you to prioritize and control the execution of jobs, ensuring efficient resource utilization and workload management. For more information about Kueue, refer to the Kueue documentation.
-
SkyPilot: SkyPilot is a framework for running AI/ML workloads, typically in a multi-cloud orchestration environment. For more information about SkyPilot, refer to the SkyPilot documentation.