Page tree
Skip to end of metadata
Go to start of metadata

For cloud-based deployments, Kyvos allows you to manage scheduling and system capacity. 

You can create a schedule to:

  • Shutdown cluster for any time interval
  • Start cluster for any time interval
  • Increase/decrease query engine capacity for any time interval

Depending on your deployment, the system modes are Running, Default, or Suspend mode. 

Scheduling

You can view and set schedules for the Kyvos cluster and query engine capacity. If you are using the Kyvos Managed Services, you will also see information about costs.

To view and manage schedules

  1. From the Toolbox, choose Setup and Schedule Cluster Scaling.
    From here, you can view the current schedule for that item. At a glance, you can see what times the cluster is running at which capacity.

    Sample schedule

  2. To change the default settings for the cluster, click Settings. These settings are applied on the cluster when no schedule is defined.
  3. From here you can: 
    • Define the default capacity at which the query engines would run when no schedule is applied
    • Configure the days for weekend and weekdays. For example, you may want to include Saturday in Weekday and define only Sunday as the weekend.
    • Change the time format between 12 or 24 hours.
  4. If you have made changes, click Save.

    Default cluster settings

To add a new schedule

  1. From the Toolbox, choose Setup and Cluster Scaling.
  2. Right-click a time in the schedule and click Add Schedule.  
    Or click Add Schedule found at the top right of the page.
  3. For Title, choose a name that provides information such as "Running with 50% query engine capacity". 
  4. Click to choose which days the event occurs and specify the time.
  5. Choose whether the service should remain online or offline.
    For example, you may want the service to be offline on the weekends at certain times.
  6. Choose a query engine capacity. 
  7. Click Add.

To edit an existing schedule

  1. From the Toolbox, choose Setup and Cluster Scaling.
  2. Right-click a schedule entry in the schedule and click Edit.  
  3. Make changes as needed and click Save.

Users will see a message when they try to sign in before a cluster goes down according to its schedule. If they are currently using the system, they will see a message that the server is scheduled to be down in 10 minutes and after that, no requests will be served. When the cluster is down, users are prompted to wait and retry the request later. 

Scaling types

Kyvos supports the following three types of cluster scaling:

  1. Count-based: In this type, you can define the number of query engines to be used in the cluster for scheduled scaling. This type is supported for all cloud platforms.


  2. Instance-type based scheduling: In this type, you can define the capacity at which the query engine instances will be used in the cluster for scheduled scaling. This type is supported for AWS clusters only.


  3. Fixed-capacity scheduling: In this type, you can define whether the query engine will run at zero or 100% capacity in the cluster for scheduled scaling. This type is supported for all cloud platforms.

You can change the type of scheduling using the QE_ELASTICITY property in the olapengine.properties from the Kyvos Manager.

Subscribe to notifications

Click Subscribe Here in the lower right of the page to subscribe to receive notifications for cluster events, such as when the cluster shuts down as per schedule, entity modifications, or license expiring soon.

You can set up notifications and subscriptions as needed.

For AWS and Azure clusters, if you subscribed to the Notify if the cluster shuts down as per schedule, you will receive an email with details on how to manually restart the cluster if you need to.

View cluster audit logs

Use the Click here link on the lower right of the page to view audit logs for cluster scaling events. This opens the Monitor page and displays all the events that have occurred on your cluster along with the details. You can view details about default and user-created schedules and download logs of any type of cluster scaling event. 

Capacity

You can adjust the capacity to accommodate peak demand times. For each event you add to the schedule, you can set a query engine capacity. Over time, as you monitor usage, you will learn when you need to increase the capacity and when you can reduce it or take the service offline. Remember to optimize cubes to improve performance which also affects capacity.

Related topics
  • No labels