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 (https://ideas.ibm.com).


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 (https://www.ibm.com/ideas) - Use this site to find out additional information and details about the IBM Ideas process and statuses.

IBM Unified Ideas Portal (https://ideas.ibm.com) - Use this site to view all of your ideas, create new ideas for any IBM product, or search for ideas across all of IBM.

ideasibm@us.ibm.com - 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 https://ideas.ibm.com


ADD A NEW IDEA

Scheduling

Showing 37

Add parameter gpumodel=same for multi node-multi gpu parallel jobs.

To allow LSF more efficient scheduling for multi node multi gpu nodes (that require to run on same type of GPU's) parameter gmodel=same is very important. Today we can only define same model for CPU type but not for GPU type: Our clients are runni...
16 days ago in Spectrum LSF / Scheduling 1 Under review

THREAD limit to be set per HOST or PER CORE instead of PER JOB

The current THREADS limit on a queue level is set per job while it has to be set per HOST or per CORE. That is because current limit doesn't reflect the size of a job so is the THREADS limit is 600 it means that a single core job can produce up-to...
29 days ago in Spectrum LSF / Scheduling 1

Enhance lsmake4 to support Grouped Target's in the latest version of make 4.4.1

Currently our tools are using Grouped Target's which don't properly work causing quite a bit of rework after the initial compile.
about 2 months ago in Spectrum LSF / Scheduling 0 Under review

Ability to reserve GPUs in LSF advanced reservations

It's currently not possible to reserve GPUs in an advanced reservation with LSF, unlike CPUs (slots) and memory. For example, we have some hosts with 120 CPU slots and 8 GPUs. I want to create a reservation that reserves half of a node. I can do t...
3 months ago in Spectrum LSF / Scheduling 0 Future consideration

Run LSF file IO operations within the container.

LSF jobs using containers depend on mounts within the container. These mounts are not available on bare metal where LSF runs file operations. The bare metal dependency means that LSF file related operations, -o, -f, -E, -Ep, etc. will not have acc...
4 months ago in Spectrum LSF / Scheduling 1

Correct on Design Flaw in Session Scheduler

Currently the Session Manager and Service Director operate in the same CGROUP on the first execution host which causes a race condition that if a task on the first execution host pushes the memory use above the memory limit (aka mem-eater process)...
5 months ago in Spectrum LSF / Scheduling 1 Not under consideration

Allow resource limit ELIGIBLE_PEND_JOBS usage for APPS

I would like to have a limit ELIGIBLE_PEND_JOBS for the consumer APPS. The reason is an application which needs licenses from a license server. In case of many simultanously dispatched jobs the license server overloads. Up to now the resource limi...
5 months ago in Spectrum LSF / Scheduling 1 Needs more information

tool to list out active and default base params - lsparams

Think of 'bparams' and the level of details that it provides with args '-la'. This helps in figuring out what all params are configured as default or what defaults values are affecting the scheduling. Such a tool does not exist for LSF Base config...
5 months ago in Spectrum LSF / Scheduling 0 Planned for future release

Group jobs to same cluster when using bsub -cluster feature

Goal: Group together lsf jobs, "regression", in such a way that bsub -cluster feature directs all jobs in the group to the same cluster. Proposal: a) bsub submits all jobs in a job array to a single cluster bsub -cluster all -J "ArrayName[1-10]" /...
6 months ago in Spectrum LSF / Scheduling 1

lsfproxyd - Persistent blocking rules

LSFProxyd is a new solution to help scheduler handle queries effectively. This RFE is to enable persistent blocksing rules for heavy query hitters. Same has been discussed with John Welch in a separate discussion.
7 months ago in Spectrum LSF / Scheduling 1 Future consideration