It’s like the cloud version of kubeflow
Vertex AI Pipelines lets you automate, monitor, and govern your machine learning (ML) systems in a serverless manner by using ML pipelines to orchestrate your ML workflows. You can batch run ML pipelines defined using the Kubeflow Pipelines (Kubeflow Pipelines) or the TensorFlow Extended (TFX) framework. To learn how to choose a framework for defining your ML pipeline, see Interfaces to define a pipeline.
difference with kubeflow
- Kubeflow uses native kubernetes PVC’s, Vertex ai uses cloud storage “fuse”
Kubeflow Pipelines and Vertex AI Pipelines handle storage differently. In Kubeflow Pipelines you can make use of Kubernetes resources such as persistent volume claims. In Vertex AI Pipelines your data is stored on Cloud Storage, and mounted into your components using Cloud Storage FUSE.
- vertex doesnt support some features of kubeflow:
- Cache Expiration: In Kubeflow Pipelines, you can specify that cached component executions expire after a specified amount of time using the Kubeflow Pipelines SDK v1 DSL.
- Recursion: In Kubeflow Pipelines, you can specify pipeline components that are called recursively.
links
references
https://cloud.google.com/vertex-ai/docs/pipelines/introduction