IBM Data and AI Ideas Portal for Customers

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:

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

Help IBM prioritize your ideas and requests

The IBM team may need your help to refine the ideas so they may ask for more information or feedback. The product management team will then decide if they can begin working on your idea. If they can start during the next development cycle, they will put the idea on the priority list. Each team at IBM works on a different schedule, where some ideas can be implemented right away, others may be placed on a different schedule.

Receive notification on the decision

Some ideas can be implemented at IBM, while others may not fit within the development plans for the product. In either case, the team will let you know as soon as possible. In some cases, we may be able to find alternatives for ideas which cannot be implemented in a reasonable time.

Additional Information

To view our roadmaps:

Reminder: This is not the place to submit defects or support needs, please use normal support channel for these cases

IBM Employees:

The correct URL for entering your ideas is:

Status Delivered
Workspace Spectrum Conductor
Components Version 2.4.1
Created by Guest
Created on Jul 1, 2020

Running only CPU based model via WMLA

We are using WMLA to run GPU based jobs fine and also able to scale across the cluster. However, we are trying to run a model only on CPU (not on GPU) using the WMLA framework. We have prepared a model which runs only on CPU and not at all sending it to GPU.

Also, while submitting a job, we did not define “gpuPerWorker” parameter. However, what we see is that the model is getting GPU slots allocated first, based on availability, and then CPU slot allocation starts once the GPU slots are over.

For Example:
Our current configuration:
Total CPU slots: 12*96
Total GPU slots: 48

Resource available before submitting the WMLA job:
Total CPU slots: 10*96
Total GPU slots: 35

We submitted a job with below params:
'maxParalleJobNum': 120

Did not define --gpuPerWorker param while submitting the job.

Our observation:
The job gets allocated to 35 GPU slots first and then starts occupying CPU slots for remaining 85 parallel tasks, the job stays in WAITING state for some time here for CPU slot allocation.