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,
Post an idea
Upvote ideas that matter most to you
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.
Normal user to monitor the resource usage of this user only
So currently following three ways are not able to meet the requirements: 1) Cluster admin can see the resource usage from explorer dashboard for all users. 2) read only roles also enable to see resource usage from all other users. 3) adding one ne...
In Explorer dashboard, we need to provide upper management information about how the cluster is being used.1 We need to see CPU slots data by user and SIG at the same time. Now if SIG is chosen for the chart, we don't see usage by users in that SI...
Inference job restarts a few times based on task exit code
During running an inference job, when a task fails and returns non-0 exit code. The inference engine should be able to retry the inference task for defined number of times should it detect non-0 exit code from the inference task.
Able to set time for model training start and duration
Before start a training job, it will be better if we can specify when the training job should be started and for how long the training job should run. DLI can have the job stopped or suspended if the job runs exceeds the specified duration.
Allow duplicate Dataset, Model names among different SIGs
In a multi-tenant environnent like ours. It is very common that different tenant will use same dataset, model for their work. Thus the chances are very high for each of them to use the same name for dataset, models. Currently when you use the same...
The Resource Group usage on Admin Dashboard is not very user friendly and requires a bit of digging to understand the metrics appropriately. Please provide an enhancement to improve the usability of this feature.
Please develop an update to allow Spectrum to read data schema information with multiple executors. Seems to only use 1 or 2 and takes a very long time compared to other YARN based environments that use multiple executors to read the data files fo...
Do not place IBM confidential, company confidential, or personal information into any field.