Skip to content

Extraction of node I/O metadata #2

@vsoch

Description

@vsoch

Not sure the extent to which ior will be used, but I want to put notes here for how we can eventually automate generation of the subsystem metadata for nodes.

  • Take in user request for names to parse from /proc/mounts or some other filtering method
  • Parse /proc/mounts to find names and get mount points (and other featresu0
  • For each mount point, we need quota (can use IOR) but...
    • Parallelism is based on the controller. We can look up device specs (and maybe find them represented somewhere online?) Aside from checking a website.

Ping @hariharan-devarajan just for future discussion - I'm going to manually create this for now so I can develop the rainbow plumbing. 🚽

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type
    No fields configured for issues without a type.

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions