Skip to Main Content
IBM Data and AI Ideas Portal for Customers

This portal is to open public enhancement requests against products and services offered by the IBM Data & AI organization. To view all of your ideas submitted to IBM, create and manage groups of Ideas, or create an idea explicitly set to be either visible by all (public) or visible only to you and IBM (private), use the IBM Unified Ideas Portal (

Shape the future of IBM!

We invite you to shape the future of IBM, including product roadmaps, by submitting ideas that matter to you the most. Here's how it works:

Search existing ideas

Start by searching and reviewing ideas and requests to enhance a product or service. Take a look at ideas others have posted, and add a comment, vote, or subscribe to updates on them if they matter to you. If you can't find what you are looking for,

Post your ideas

Post ideas and requests to enhance a product or service. Take a look at ideas others have posted and upvote them if they matter to you,

  1. Post an idea

  2. Upvote ideas that matter most to you

  3. Get feedback from the IBM team to refine your idea

Specific links you will want to bookmark for future use

Welcome to the IBM Ideas Portal ( - Use this site to find out additional information and details about the IBM Ideas process and statuses.

IBM Unified Ideas Portal ( - Use this site to view all of your ideas, create new ideas for any IBM product, or search for ideas across all of IBM. - Use this email to suggest enhancements to the Ideas process or request help from IBM for submitting your Ideas.

IBM Data & AI Roadmaps ( - Use this site to view roadmaps for Data & AI products.

IBM Employees should enter Ideas at

Status Future consideration
Workspace Spectrum LSF
Components Scheduling
Created by Guest
Created on Jun 13, 2022

Preemption support for MIG GPU jobs

We would like that preemption will be supported also for MIG jobs(A100 dgx mig devices). Currently the preemption is not working for this type of jobs.

Example scenatiois:

kill two or more jobs and reconfigure mig to have the new job run. We need to consider which slices are currently in use, and which could be combined etc. For example, consider the following cases on A100 40GB:

Case 1: Say slices 0 2 3 are in use, with 1 4 5 6 being idle.

If the user asks for half the memory, then it can fit in 4 5 6

But if the user asks for 4 slices (which is also half the memory) then the jobs on 0 2 3 would need to be killed.

Case 2: Say slices 0 1 2 5 6 are in use, with 3 and 4 idle

The user asks for 2 slices or 1/4 of the memory, though 3 and 4 are idle, slides 3 and 4 cannot be combined, we'd have to kill 0 1 2 or 5 6 to allow this job to run.

Needed By Month
  • Guest
    Feb 22, 2023

    Hello, are there any news regarding this feature request? we are already in the future :)

  • Admin
    Bill McMillan
    Aug 8, 2022

    This is something we may consider in the future.