Skip to content

[Initiative]: Cloud Native AI Sustainability Papers #1675

@zanetworker

Description

@zanetworker

Name

Cloud Native AI & Sustainability

Short description

Two white papers for the AI sustainability

Responsible group

TOC

Does the initiative belong to a subproject?

Yes

Subproject name

Cloud Native AI Working Group

Primary contact

@zanetworker

Additional contacts

@raravena80

Initiative description

Paper 1: The paper aims to provide a framework and recommendations for building and managing AI systems with sustainability in mind, emphasizing resource optimization and reducing the environmental footprint.

Paper 2: The paper discusses the challenges and opportunities for making Generative AI more sustainable in cloud-native systems. It highlights the high energy consumption of AI training and inference, issues like inefficient GPU usage and data management, and proposes strategies for optimization. Key areas of focus include energy tracking, power capping, efficient resource scheduling, and the use of tools like Kepler, Kueue, and KEDA to manage AI workloads more sustainably.

Original issue: cncf/tag-runtime#168

Source: https://docs.google.com/document/d/1tNfR5yxnqfu9fduEb3108oBGXQ6spyRzA6cBTs9zbnQ/edit?tab=t.fwj647zi37u1#heading=h.a2csx0fpflu3

Deliverable(s) or exit criteria

Final draft versions for both papers to be handed off to the CNCF publishing staff.

Metadata

Metadata

Assignees

No one assigned

    Labels

    kind/initiativeAn initiative or an item related to imitative processes

    Type

    No type

    Projects

    Status

    New

    Status

    status/new

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions