Change a cluster from reserved to elastic mode
Change a cluster from reserved mode to elastic mode to reduce costs with tiered storage of hot and cold data and time-based scaling, and to isolate computing resources for different resource groups.
Prerequisites
-
Your Alibaba Cloud account has no unpaid renewal orders.
-
Only clusters with minor version 3.1.3.9 or later support changing specifications across different instance families.
NoteTo view and update the minor version, go to the Configuration Information section on the Cluster Information page in the AnalyticDB for MySQL console.
-
The cluster is in the Running state.
Usage notes
-
Changing from reserved mode to elastic mode involves provisioning resources with new specifications, performing full and incremental data migration, and switching to the new cluster. The duration depends on the cluster specifications and data volume.
-
You cannot change other configurations while changing the ESSD performance level.
-
During the switchover, the original cluster enters a read-only state for 5 to 10 minutes.
You can connect to the original cluster and run the following statement to specify a switchover time window. The following example sets the window to 23:00–23:30.
SET adb_config REPLICATION_SWITCH_TIME_RANGE=`23:00, 23:30`;For information about how to connect to a cluster, see Connect to a cluster.
-
When you change a C32 cluster from reserved mode to elastic mode, the core count of a single node is reduced to 8. By default, a BUILD job uses 3 cores, which can increase the average CPU utilization. If the increased CPU utilization does not affect your services, you can ignore it. If your services are affected, scale up your cluster or Submit a ticket for technical support. For more information about BUILD jobs, see BUILD.
Billing
For more information, see Change the billing method of a cluster.
Procedure
-
Log on to the AnalyticDB for MySQL console. In the upper-left corner, select the cluster's region. In the left-side navigation pane, click **Clusters**. On the **Data Warehouse Edition** tab, find the target cluster and click Scale Up or Scale Down in the **Actions** column.
-
On the Upgrade/Downgrade page, select Elastic Mode and configure the parameters.
Parameter
Description
Edition
Fixed to Data Warehouse Edition. No selection required.
Edition
Select Cluster Edition.
Specifications
Fixed as E32 and cannot be changed.
Computing Resources
Select the computing resource for the cluster. Computing resources are used for data computing. Increasing computing resources can improve data query speed. We recommend that the CPU cores of computing resources are more than those of elastic I/O resources.
EIU Capacity
Select the elastic I/O resource specification for the cluster. The specification selected here is for a single node. Each group of elastic I/O resources contains 3 nodes.
-
When the single-node specification is 8 cores and 64 GB, the specification of one group of elastic I/O resources is 24 cores and 192 GB.
-
When the single-node specification is 12 cores and 96 GB, the specification of one group of elastic I/O resources is 36 cores and 288 GB.
-
When the single-node specification is 16 cores and 128 GB, the specification of one group of elastic I/O resources is 48 cores and 384 GB.
Elastic I/O Resource (Elastic IO Unit, EIU for short) is the base unit for measuring cluster storage performance in elastic mode. You can use EIUs to independently scale out storage resources. For more information, see Scale out elastic I/O resources (EIU).
After elastic I/O resources are scaled out, the amount of hot data locked by each elastic I/O resource increases. For more information, see Storage performance metrics based on EIUs. If you need to scale back, make sure that the hot data volume is less than the maximum hot data volume that can be locked by the scaled-in specification. If the hot data volume exceeds the maximum, we recommend that you move some data to cold storage before you perform the scale-in operation.
ImportantYou can change the elastic I/O resource specification only for clusters that meet the following conditions:
-
The mode is Elastic Mode and the Computing Resource is 16 cores 64 GB (Cluster Edition), 24 cores 96 GB (Cluster Edition), 32 cores 128 GB, or a higher specification.
-
The cluster is a new pricing cluster, which means a cluster created in a mainland China region on or after September 1, 2022.
EIUs
Select the number of elastic I/O resource groups.
The maximum number of groups you can purchase varies by elastic I/O resource specification.
-
If the elastic I/O resource specification is 8 cores and 64 GB, you can purchase up to 32 groups.
-
If the elastic I/O resource specification is 12 cores and 96 GB or 16 cores and 128 GB, you can purchase up to 16 groups.
NoteThis configuration is supported only when the mode is Elastic Mode and the Computing Resource is 16 cores 64 GB (Cluster Edition), 24 cores 96 GB (Cluster Edition), 32 cores 128 GB, or a higher specification.
ESSD Performance Level
Select the cloud disk level for the cluster. The following list describes the capacity ranges for different cloud disk types:
-
PL0: 20 to 65536 GiB.
-
PL1: 20 to 65536 GiB.
Important-
When you change the ESSD performance level, you cannot change other configurations.
-
The amount of hot data must meet the minimum capacity requirement of the selected performance level when you change the ESSD performance level.
-
Changing the ESSD performance level affects the cost of hot storage space. The cost is calculated as:
Unit price of storage space for the ESSD performance level * Hot storage usage * Duration. For unit prices, see Data Warehouse Edition Product Pricing.
-
-
For a Subscription cluster, read and accept the Terms of Service, click Buy Now, confirm the order details and payment method, and then click Subscribe. For a Pay-as-you-go cluster, read and accept the Terms of Service, and then click Buy Now.