Hologres:Scheduled scaling (beta)

Regional availability

Scheduled scaling is in public preview. To apply for trial access, use your Alibaba Cloud main account to complete the Hologres scheduled scaling public preview application.

Go to Virtual Warehouse Management

1. Log on to the Hologres console. In the top menu bar, select the target region.

2. In the left navigation pane, click Instances, then click the target Instance ID.

3. On the instance details page, click Virtual Warehouse Management in the left navigation pane.

View and manage compute groups

On the Virtual Warehouse Resource Management tab, you can see allocated and unallocated reserved CU and elastic CU across all compute groups.

• To create a compute group, click Create Virtual Warehouse. In the dialog box, enter a Virtual Warehouse Name and specify the Virtual Warehouse Resource.

• For a compute group with a Status of Running, you can Modify Configuration, Restart, Stop, or Rebalance it. For details on Rebalance, see Shard rebalancing (Rebalance).

The default init_warehouse compute group cannot be stopped.

View the elasticity plan timeline

1. On the Virtual Warehouse Management page, click the Virtual Warehouse Scaling Plan tab.

2. From the Instance/Virtual Warehouse drop-down list, select an instance or compute group to see its upcoming scaling schedule. The timeline shows planned future operations, not historical metrics.

Add or edit an elasticity plan

On the Virtual Warehouse Scaling Plan tab, click the icon to the left of the target compute group, then click Add Time Segment. Set the start time, end time, and elastic resource amount for the daily period. Click Save in the Actions column to apply.

To update an existing plan, click Edit. To remove it, click Delete.

How changes take effect immediately

When you save a plan, Hologres checks whether the change affects the current time window. If it does, the new configuration applies right away.

Example 1 — Adding a plan that covers the current time

Current time: 16:00. Compute group A has 64 reserved CU.

You add a plan: effective period 15:00–18:00, elastic resources 32 CU.

Result: Compute group A immediately scales out to 96 CU total.

Example 2 — Narrowing a plan so the current time falls outside it

Current time: 16:00. Compute group A has 64 reserved CU with an existing plan:

• Before: effective period 15:00–18:00, elastic resources 32 CU (currently applied; total = 96 CU)

• After: effective period 17:00–18:00, elastic resources 32 CU

Result: Compute group A immediately scales in from 96 CU to 64 CU.

Scaling plan execution logs

1. On the Virtual Warehouse Management page, click the Elastic Event Execution Logs tab.

2. Select a time range to review past scaling executions. The log shows execution time, compute group name, status, event type, reserved CU, and target elastic CU.

Monitoring metrics

The Warehouse_timed_elastic_cores(Count) metric in the Hologres console shows the number of cores added by scheduled scaling. Configure alert rules for this metric as needed. For details, see Monitoring metrics in the Hologres console.

Cloud Monitor events

All scaling operations triggered by elasticity plans are recorded in Cloud Monitor.

View scaling events:

1. Go to Cloud Monitor Event Center. On the System events page, under Event monitoring, select Hologres.

2. Look for the following event names:

   • Instance:Warehouse:TimedElastic:Start — scaling started

   • Instance:Warehouse:TimedElastic:Finish — scaling completed

   • Instance:Warehouse:TimedElastic:Failed — scaling failed

Set up alerts:

Configure notifications based on these events. For details, see Use system events for alerts.

Example failure event payload:

{
    "Status": "Failed",
    "InstanceName": "<instance_id>",
    "ResourceId": "<instance_resource_id>",
    "Content": {
        "ScaleType": "ScaleDown",       // "ScaleUp" or "ScaleDown"
        "ScheduleId": "xxxxxx",         // Elasticity plan ID
        "TimedElasticCPU": 0,           // Target elastic CPU cores after scaling
        "WarehouseId": "2",             // Compute group ID
        "WarehouseName": "<warehouse_name>"
    },
    "Product": "hologres",
    "Time": 1722852008000,              // Unix timestamp in milliseconds
    "Level": "WARN",
    "RegionId": "<region>",
    "Id": "<event_id>",
    "GroupId": "0",
    "Name": "Instance:Warehouse:TimedElastic:Failed"
}

Key fields in the failure event:

ActionTrail

Console operations such as editing elasticity plans, and the execution of scaling operations, are recorded in ActionTrail. For details, see Event audit logs.