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 updateson 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,
Post an idea
Upvote ideas that matter most to you
Get feedback from the IBM team to refine your idea
Specific links you will want to bookmark for future use
RTM should be able to properly parse effective and/or combined resource requirements and handle both compound and /task, /host, and /job reservations to determine pending/reserved memory
RTM calculates pending and running memory reservations by parsing the resource requirements string. However, it currently is not properly calculating that memory reservation for either compound resource requirements or with the new per /task, /hos...
For operations and KPI management, having some additional metrics in the various stats tables is important. For our use here, it would be helpful to have the following metrics: Pend Time Metrics: (eligible based upon job class not LSF setting)* av...
Add a JobID's column to the table grid_hosts_gpu and show in Job Details page
It would be very beneficial to reporting to see which jobid's are allocated to which jobs when it comes to the GPU assigned. This way, the user can look at the ELIM graphs and know that their job is actually using the GPU in question. Right now, y...
Add additional Metrics to the grid_clusters_perfmon_status table
From badmin show status add the following missing columns:
Latest estimation start: -
Next estimation start: -
Estimation status: Not started
Active estimator PIDs: -
Elapsed time: 0:00:00
Number of jobs to estimate: 0
Number of estimated jobs:...
Add Resource Description to Table grid_hosts_resources
We currently have a need to view Shared Resources via their Description. It would be nice if:
1) This data was in the grid_hosts_resources table
2) There was a searchable description column under the Shared Resources Report view in RTM
Ability to limit user's views in RTM to select clusters
Reduces support because users can't flip to other clusters via Cluster drop-down menu to then ask questions about resources and usage or see other information that should only be for users of a particular cluster or why can't they find their infor...
Ability to specify cores or exclusive host for Benchmark job
Currently all benchmark jobs submitted via RTM use 1 core. Also, one cannot specify node to be exclusively used via bsub -x. This allows other jobs to land on same node and thus tainting benchmark numbers
Move away from having to set job_end_logged on every gridjobs and gridpend poller runs
The setting of job_end_logged to 0 every polling cycle can add as much as 100 seconds to every gridjobs or gridpend run. If instead, the poller were to capture the following
SELECT MAX(last_updated) FROM grid_jobs* WHERE clusterid = ?
Provide "bkill -C 'some comment' jobid" Comment Message in RTM's grid_jobs and grid_jobs_finished tables
We will be doing automated killing of workload in 2022, and now that LSF provides the 'bkill -C' option, we would like that Kill message to be stored in RTM for auditing the reasons why jobs were killed by automation.
Do not place IBM confidential, company confidential, or personal information into any field.