Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Metrics support enhancements: 1) Splitting up amd_gpu_memory_use_perc… #5

Open
wants to merge 1 commit into
base: master
Choose a base branch
from

Conversation

japarada
Copy link
Contributor

…ent into two separate metrics: amd_gpu_memory_used and amd_gpu_memory_total. 2) Adding 'gpu' label which is the index of the gpu on the node. 3) Adding 'node' label which is the name of the node on which the gpu runs. 4) Adding pod name label.

…ent into two separate metrics: amd_gpu_memory_used and amd_gpu_memory_total. 2) Adding 'gpu' label which is the index of the gpu on the node. 3) Adding 'node' label which is the name of the node on which the gpu runs. 4) Adding pod name label.
nodename, err := exec.Command("uname", "-n").Output()
if err == nil {
gNodeName = string(nodename)
gPod = string(nodename)

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Why do we need gPod and gNodeName - what is gPod' supposed to be, is it always same as gNodeName?

nil,// The metric's constant label dimensions.
),
GPUMemoryUsage: prometheus.NewDesc(
GPUMemoryBusyPercent: prometheus.NewDesc(

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

By "busy" memory, does it mean "used"? Not sure what 'busy' means for memory? Could we changed to "used"?

Copy link
Contributor

@muralimk-amd muralimk-amd left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

When we try to pass arguments "node_id" and "pod" in the query, we are getting "Empty Query result".

@@ -141,52 +143,63 @@ func NewCollector(handle func() (collect.AMDParams)) prometheus.Collector {
GPUDevId: prometheus.NewDesc(
prometheus.BuildFQName("amd", "", "gpu_dev_id"),
"AMD Params",// The metric's help text.
[]string{"gpu_dev_id", "productname"},// The metric's variable label dimensions.
[]string{"gpu_dev_id", "productname", "node_id", "pod"},// The metric's variable label dimensions.
Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

argument "pod" is not working for any of the APIs when we try to query. Please check
image

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Query with argument "node_id" also gives "Empty Query result"
image

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants