Today we’re announcing improved Snowflake Task-based orchestration for containeriazed Jobs for ML Jobs and Snowpark Container Services customers
Yavor is a PM at Snowflake working on developer experience. Previously at Docker, Auth0, Hulu, and Microsoft Azure.
Today we’re announcing improved Snowflake Task-based orchestration for containeriazed Jobs for ML Jobs and Snowpark Container Services customers
Today we announce the General Availability of the Ingress Gateway, providing advanced traffic management capabilities in SPCS. This feature enables stable URLs and traffic splits as part of a mature deployment strategy. Combined with compute support for placement groups, this feature also offers a complete toolkit for building mission-critical, high-availability applications.
We're now making it easier than ever to build multi-user containerized apps that leverage Snowflake's RBAC model. Check out this example of a Gradio chatbot using Cortex Analyst, while using the user/caller's rights.
When deploying containers in Snowflake via Snowpark Container Services (SPCS), the ability to debug container failure conditions is key as workloads mature and go to production. To address these scenarios, SPCS infrastructure now emits Platform Events around container status.
Let's walk through an example of how to use some awesome new Snowflake Trail observability capabilities, which make developing, debugging, and monitoring SPCS applications a breeze
11 March 2025
As part of Snowpark Container Services (SPCS), Snowflake offers a Jobs concept, which is well-suited for some important workloads in AI/ML and Data Engineering. By using the optional asynchronous Job processing support, users can save time and reduce cost with long-running workloads.