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 Employees should enter Ideas at

Status Not under consideration
Workspace Spectrum LSF
Created by Guest
Created on Aug 10, 2022

bsub should accept a cpu_efficiency parameter to allow cpu over-commiting in the pool

It's rare for a job to achieve close to 100% cpu efficiency. Often, jobs have a peak cpu requirement during its execution lifetime but use less cpus in average. If bsub could be told that a job is expected to run with a given cpu efficiency, LSF could allocate multiple jobs with sufficiently-low cpu efficiency on the same execution host even if the total number of cpus requested by all jobs exceed that in the host. This is what I'm calling cpu over-committing. The idea here is that it is unlikely that all jobs will peak at the same time. CPU over-committing will enhance the entire pool efficiency. This feature could be enabled/disabled either globally or perhaps down to the queue level. One or more parameters may be required to tell LSF how to allocate these jobs. Perhaps a maximum execution cpu over-commit percentage (should be greater than 100% to have any cpu over-commit), or a maximum cpu_efficiency for a job to be considered for cpu over-commit.
Needed By Yesterday (Let's go already!)
  • Admin
    Bill McMillan
    Sep 29, 2022

    The degree of over commitment is really application dependent, and not something that could generically be applied to a host.

    Probably the simplest approach would be to scale the requested resources by efficiency factors

    bsub -a "eff(mem=0.75, cpu=0.5)" -n 20 -R "rusage[mem=10000]" a.out

    and the eff.esub modifies it to -n 10 -R "rusage[mem=7500]"

  • Guest
    Aug 22, 2022

    I agree with the above and think it should also handle memory_efficiency to allow for memory over-committing.