Skip to content

Commit

Permalink
fix: update the volcano metric document.
Browse files Browse the repository at this point in the history 
Signed-off-by: tanjie.master <tanjiemaster@gmail.com>
  • Loading branch information
@fengruotj
fengruotj authored and tanjie.master committed Oct 21, 2024
1 parent 4773047 commit dbdfc8e
Showing 1 changed file with 25 additions and 24 deletions.
49 changes: 25 additions & 24 deletions docs/design/metrics.md
View file
Original file line number Diff line number Diff line change
@@ -1,39 +1,40 @@
## Scheduler Monitoring

## Introduction
Currently users can leverage controller logs and job events to monitor scheduler. While useful for debugging, none of this options is particularly practical for monitoring kube-batch behaviour over time. There's also requirement like to monitor kube-batch in one view to resolve critical performance issue in time [#427](https://github.com/kubernetes-sigs/kube-batch/issues/427).
Currently users can leverage controller logs and job events to monitor scheduler. While useful for debugging, none of this options is particularly practical for monitoring volcano behaviour over time. There's also requirement like to monitor volcano in one view to resolve critical performance issue in time [#427](https://github.com/kubernetes-sigs/volcano/issues/427).

This document describes metrics we want to add into kube-batch to better monitor performance.
This document describes metrics we want to add into volcano to better monitor performance.

## Metrics
In order to support metrics, kube-batch needs to expose a metrics endpoint which can provide golang process metrics like number of goroutines, gc duration, cpu and memory usage, etc as well as kube-batch custom metrics related to time taken by plugins or actions.
In order to support metrics, volcano needs to expose a metrics endpoint which can provide golang process metrics like number of goroutines, gc duration, cpu and memory usage, etc as well as volcano custom metrics related to time taken by plugins or actions.

All the metrics are prefixed with `kube_batch_`.

### kube-batch execution
This metrics track execution of plugins and actions of kube-batch loop.
### volcano execution
This metrics track execution of plugins and actions of volcano loop.

| Metric name | Metric type | Labels | Description |
| ----------- | ----------- | ------ | ----------- |
| e2e_scheduling_latency | histogram | | E2e scheduling latency in seconds |
| plugin_latency | histogram | `plugin`=&lt;plugin_name&gt; | Schedule latency for plugin |
| action_latency | histogram | `action`=&lt;action_name&gt; | Schedule latency for action |
| task_latency | histogram | `job`=&lt;job_id&gt; `task`=&lt;task_id&gt; | Schedule latency for each task |
| **Metric Name** | **Metric Type** | **Labels** | **Description** |
|---------------------------------------------|-----------------|------------------------------------------------------------------------------------------|--------------------------------------------------------------------------------|
| `e2e_scheduling_latency_milliseconds` | Histogram | None | End-to-end scheduling latency in milliseconds (scheduling algorithm + binding) |
| `e2e_job_scheduling_latency_milliseconds` | Histogram | None | End-to-end job scheduling latency in milliseconds |
| `e2e_job_scheduling_duration` | Gauge | `job_name`=&lt;job_name&gt;, `queue`=&lt;queue&gt;, `job_namespace`=&lt;job_namespace&gt;| End-to-end job scheduling duration |
| `e2e_job_scheduling_start_time` | Gauge | `job_name`=&lt;job_name&gt;, `queue`=&lt;queue&gt;, `job_namespace`=&lt;job_namespace&gt;| End-to-end job scheduling start time |
| `plugin_scheduling_latency_milliseconds` | Histogram | `plugin`=&lt;plugin_name&gt;, `OnSession`=&lt;OnSession&gt; | Plugin scheduling latency in milliseconds |
| `action_scheduling_latency_milliseconds` | Histogram | `action`=&lt;action_name&gt; | Action scheduling latency in milliseconds |
| `task_scheduling_latency_milliseconds` | Histogram | None | Task scheduling latency in milliseconds |


### kube-batch operations
This metrics describe internal state of kube-batch.
### volcano operations
This metrics describe internal state of volcano.

| Metric name | Metric type | Labels | Description |
| ----------- | ----------- | ------ | ----------- |
| pod_schedule_errors | Counter | | The number of kube-batch failed due to an error |
| pod_schedule_successes | Counter | | The number of kube-batch success in scheduling a job |
| pod_preemption_victims | Counter | | Number of selected preemption victims |
| total_preemption_attempts | Counter | | Total preemption attempts in the cluster till now |
| unschedule_task_count | Counter | `job`=&lt;job_id&gt; | The number of tasks failed to schedule |
| unschedule_job_counts | Counter | | The number of job failed to schedule in each iteration |
| job_retry_counts | Counter | `job`=&lt;job_id&gt; | The number of retry times of one job |
| **Metric Name** | **Metric Type** | **Labels** | **Description** |
|-------------------------------|-----------------|----------------------------------|------------------------------------------------------|
| `schedule_attempts_total` | Counter | `result`=&lt;result&gt; | The number of attempts to schedule pods |
| `pod_preemption_victims` | Gauge | None | The number of selected preemption victims |
| `total_preemption_attempts` | Counter | None | Total preemption attempts in the cluster till now |
| `unschedule_task_count` | Gauge | `job_id`=&lt;job_id&gt; | The number of tasks failed to schedule |
| `unschedule_job_counts` | Gauge | None | The number of jobs could not be scheduled |


### kube-batch Liveness
Healthcheck last time of kube-batch activity and timeout
### volcano Liveness
Healthcheck last time of volcano activity and timeout

0 comments on commit dbdfc8e

Please sign in to comment.