Skip to content

GPU/CPU metrics need to be more specific for each future #8

Description

@Saaketh0

Currently, the CPU/GPU data in the requests being sent to the database are being pulled straight from the .yaml config file. So if the resources are not being used fully, it won't show.

Need to implement more precise tracking of these two metrics during a job, and then populate the redis futures file with that data.

Currently in ventis/controllers/utils/sqlalchemy.py, on around line 62 the GPU/CPU is being pulled from the config file as the variable res.

Metadata

Metadata

Labels

enhancementNew feature or request

Type

Projects

No projects

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions