Skip to main content

Automatic Right-Sizing Recommendations and Rules

Cloud service provider relevance: EKS and AKS

This topic shows you how to view right-sizing recommendations for workloads and containers and work with right-sizing rules.

Before you begin, see Ocean Cluster Automatic Right-Sizing for a full description of this Ocean feature.

To view your right-sizing recommendations and rules:

  1. In the left main menu, click Ocean > Cloud Clusters.
  2. Select a cluster from the list of clusters.
  3. Click Right Sizing > Advanced Optimization.

The Advanced Optimization tab contains these lists:

  • Workloads Optimization List.
  • Automation Rules List.

Your workload optimization activities impact the status of the workloads in the Right Sizing Savings panel

Workloads Optimization List​

This list displays your right-sizing recommendations per workload and lets you drill down per container.

  • Right Sizing rules that are attached to specific workloads.
  • Workload optimization status: If the workload is attached to a right-sizing rule, the name of the rule appears under the Rules column. The rule can have one of the following workload (colored) optimization statuses:
    • Green (Optimized): Workload is fully optimized, and no action is required.
    • Orange (Limited): Ocean applied recommendations to the workload, but there are limitations due to overhead, HPA, etc.
    • Blue (Pending): Rule has been attached to the workload but is out of schedule.
    • Red (Not Optimized): Issues due to user configuration, for example, thresholds or exclude preliminary recommendations.
    • Gray (Rollback): Ocean rolled back to the original deployment request and suspended the workload's attachment to the rule.
    • Brown (Not Attached): The Workload is not optimized.
  • Workload type and names.
  • vCPU and memory right sizing recommendations per deployment. Recommended increases are shown with a green up arrow, and recommended decreases are shown with a red Down arrow.
  • HPA: If the workload is configured with HPA, ON is displayed under HPA. Hover over the entry for information about the specific HPA trigger (CPU/Memory/other).
  • Potential monthly maximums savings if you adopt the recommendations.
note
  • Hover over the Limited and Not optimized statuses to view more details in a tooltip.
  • There are no vCPU / memory recommendations or Potential monthly max if a workload is fully optimized. Savings are displayed for the workload because Ocean is already optimizing it.
  • If the button appears on the right of the screen, workloads were moved to rollback status. See Acknowledge a Workload Rollback

If a message about VPA not reporting is displayed:

  • You cannot attach rules to workloads.
  • If at least 1 rule was previously attached to a workload, the workload is not optimized due to the missing VPA installation in the cluster.

The graphical display above the list shows the breakdown of these workload optimization statuses.

To view a list of your potential savings and recommendations per container:

  • Click on the down arrow to the left of a workload to drill down to the containers. For each container, you can then view the following:

    • vCPU Request: showing current and average utilization and a recommended increase or decrease for this resource (in vCPU units). If no changes are required, a Keep icon is displayed.
    • Memory Request: This shows current and Average utilization and a recommended increase or decrease for this resource (in MiB units). If no changes are required, a Keep icon is displayed.
    • Right-Sizing Recommendations: Show the recommended changes in vCPU and memory. Click on the Copy icon to save these changes for later.

Automation Rules List​

automation-rules-e

This list displays your existing right-sizing rules.
Each rule entry shows relevant information about the parameters that trigger the rule and its scheduling plan.

Work with Right Sizing Rules​

You can create right-sizing rules to trigger immediately after a specific set of requirements is met or at a specific time after the requirements are met.

Create or Edit a Right-Sizing Rule​

To create/edit a right-sizing rule:

  1. Click the Advanced Optimization tab if not already displayed.

  2. To create a new rule, click + Add new rule above the Automation Rules list (or to edit an existing rule, click the pencil icon in the rule).

  3. In the Configure Automation Rule dialog box, enter/edit the unique rule name.

  4. Select when to apply the recommendation by selecting one of the following options:

    • Once available: The recommendation is applied immediately after it becomes available.
    • Specific time: You select when to apply the recommendation after it becomes available.

    rule-when-to-apply-3

  5. Turn on Exclude preliminary recommendation if you want to suppress recommendations as long as the workload has preliminary status (4 days).

  6. Select one of the Restart replicas options:

    • All manifests.
    • Manifests with more than 1 replica only.
    • No restart.

  7. Click the Set the resources percentage change down arrow to apply the recommendation, and set the CPU and Memory percentage thresholds. This is the minimum percentage change from the current request for applying a recommendation. If the right-sizing recommendation exceeds the percentage threshold for either resource (CPU or Memory), it will be applied to both resources, and the resulting status will be fully optimized. We do this because the original purpose of the threshold is to prevent unnecessary pod deletion. However, if we need to delete a pod and relaunch a new one for one resource, we do the same for the other.

  8. Click the Set recommendation ranges for resources down arrow and enter the upper and lower boundary values for CPU (millicpu) and Memory (MiB) requests to apply a recommendation. By default, the minimum values are 10 millicpu for CPU and 32 MiB for memory; no lower values will be accepted.

    • If a recommendation is above the set boundaries, automatic right-sizing will apply the recommendation using the maximum value configured in the rule.
    • If a recommendation is below the set boundaries, automatic right-sizing will apply the recommendation using the minimum value configured in the rule.

  9. Click the Set overhead for resources down arrow and set the CPU and memory percentage overheads. An overhead specifies the percentage of extra resources to add to the new request recommendation.

  10. Ocean supports automatic right-sizing for HPA-associated workloads. To enable, click Apply HPA on associated workload.

  11. Turn on Auto-attach if you want to automatically attach rules to workloads based on selected criteria.

    • In the Auto-attach area, select required namespaces / labels.

  12. After you save the rule, it appears in the area under the Workloads Optimization list.

    note
    • Default values for Overhead and Automation Threshold are 10% and 5% respectively.
    • The 10% default overhead is calculated within the recommendation itself.
    • Threshold value is only used for down-sizing cases

Attach a Right-Sizing Rule to One or More Workloads​

Options:

  • Auto-attach: Turn on auto-attach and let Ocean automatically attach rules to workloads based on selected criteria. See Create or Edit a Right-Sizing Rule.

    note

    Auto-attach attaches rules to existing and newly-added workloads.

  • Manual attach: Manually attach a specific right-sizing rule to one or more workloads.

    To manually attach a rule:

    1. Select one or more workloads in the Workloads Optimization list.

    2. From the Actions drop-down menu above the table, click Attach Rule.

      attach-rule-to-workload

    3. You can either attach an existing or new rule you create from scratch (a new rule will be attached to the workload(s) you selected earlier):

    4. Save and apply the rule.

Detach a Right-Sizing Rule from One or More Workloads​

To detach a rule from one or more workloads:

  1. Select workloads in the Workloads Optimization list.
  2. From the Actions drop-down menu above the list, click Detach Rule.
important

If you encounter Kubernetes issues, we recommend detaching workloads from rules and rolling them in your cluster.

Delete a Right-Sizing Rule​

To delete a right sizing rule:

  1. To the right of the row for the rule in the list, click the wastebasket icon.
  2. When the confirmation message appears, Click Delete, or Cancel (if you are unsure).
important

You cannot restore a deleted right-sizing rule. In addition, a rule may be deleted only if it is no longer attached to a workload.

Acknowledge a Workload Rollback​

If a workload encounters an OOM error, Ocean rolls back to the original deployment request and suspends the workload's attachment to the rule. The workload moves to rollback status. When at least one workload has rollback status, the button appears at the top-right of the screen.

To acknowledge a workload rollback:

  1. Click Acknowledge Rollback to view all the workloads with the rollback status.

    right-sozomg-rollback-dialog

    • The rollback drill-down list contains the following information:
      • Workload Name.
      • Namespace.
      • CPU Update in vCPUs (before and after rollback).
      • Memory Update in MiBs (before and after update).
      • Rollback Time: In format MM/DD/YYYY HH:MM:SS AM/PM

  2. Select the checkboxes for the required workloads.

  3. Click I Acknowledge the Rollback.

The workloads are displayed in the Workloads Optimization List without any attached rules. Before attaching a rule to a rolled-back workload, first fix the issue.

Set the vCPU/Memory Percentile​

You can select the right-sizing percentile settings to calculate the vCPU and memory recommendations. The lower the percentile, the stronger the recommendations.

By default:

  • vCPU: Right-sizing uses the 85th percentile.
  • Memory: Right-sizing uses the maximum value.
important

Changing the percentile setting will impact the recommendations that were already applied (this may take a few minutes to update).

To change settings:

  1. Click Settings above the workloads optimization list.

  2. Click the arrow on the right for vCPU or Memory as required (vCPU shown in the example).

  3. Change the current value(s) and save.

Best Practices​

These are the Right-Sizing Best Practices:

  • Workload limits should not have the same values as requests.
  • If you set overheads for resources, start with a relatively high overhead (20%) and decrease it with time.
  • If you set boundaries (recommendation ranges for resources), avoid applying the specific rule to all workloads. All services have different purposes.