ApsaraDB for ClickHouse:Tiered storage of hot data and cold data

Prerequisites

Background information

• Hot data is the data that is frequently accessed and stored on enhanced SSDs (ESSDs) or ultra disks. These disks are selected when you create clusters. Hot data storage is suitable for business scenarios that require high-performance queries.

• Cold data is the data that is not frequently accessed. Cold data is stored on cold data disks at low storage costs to meet the requirement of cost-effective storage.

  Tiered storage of hot data and cold data provides the following storage policies.

  Storage policy	Description
  Default storage policy	Newly written data is stored on hot data disks to support efficient queries. If the amount of stored hot data reaches the storage threshold, the part data file whose size is the largest in the hot data disk is automatically migrated to a cold data disk. This way, newly written hot data can be stored on hot data disks.
  Storage policy based on time to live (TTL)	Add a TTL clause to the default storage policy. This way, all data generated before the specified period in the TTL clause is automatically migrated to a cold data disk.

Precautions

• If you enabled the data backup feature for an ApsaraDB for ClickHouse cluster, you cannot enable tiered storage of hot data and cold data for this cluster.

• When recommend that you enable tiered storage of hot data and cold data only during off-peak hours. Enabling tiered storage of hot data and cold data causes the restart of the cluster. During the restart, you can only view the cluster information. You cannot upgrade, scale out, or scale in the cluster, or modify parameter configurations.

• When you use tiered storage of hot data and cold data, you must specify storage policies for tables.

  • When you create a table, you can use SETTINGS storage_policy = 'hot_to_cold' to specify the data storage policy.

  • You can modify the data storage policy of an existing table.

    Clusters with tiered storage of hot data and cold data enabled after January 1, 2024

    Execute the following statement to modify the data storage policy:

    ALTER TABLE <table_name> MODIFY SETTING storage_policy = 'hot_to_cold';

    The following table describes the parameters in the preceding statement.

    Parameter	Description
    table_name	The name of the table.

    Clusters with tiered storage of hot data and cold data enabled before January 1, 2024

    You can submit a ticket to apply for adding a hot and cold data storage policy. This way, you can use the policy to migrate data from a hot data disk to a cold data disk.

• The tiered storage of hot data and cold data feature cannot be disabled after you enable this feature.

Billing

• You are billed for the tiered storage of hot data and cold data feature based on the following rules.

  Storage type	Billing details
  Hot data storage	The storage space that you purchase for a cluster is the storage space for hot data. You do not need to purchase storage for hot data after you purchase a cluster.
  Cold data storage	You are billed for cold data storage based on the pay-as-you-go billing method. You do not need to purchase storage for cold data before you use the storage. You are billed for cold data storage on an hourly basis based on the used storage.

• For more information about the unit prices of cold data storage, see Storage fees of ApsaraDB for ClickHouse clusters.

• Billing formula: Fee of cold storage space = Unit price of cold data storage × Used cold data storage × Usage duration.

• For example, if the unit price of cold data storage is USD 0.000026 per GB-hour and 2,000 GB of cold data is stored from 19:00 to 20:00, the total fee is calculated based on the following formula: USD 0.000026 per GB-hour × 2,000 GB × 1 hour = USD 0.052. If you release 500 GB of storage space for cold data between 20:00 and 21:00, only 1,500 GB storage is used. In this case, the total fee incurred between 20:00 and 21:00 is calculated based on the following formula: USD 0.000026 per GB-hour × 1,500 GB × 1 hour = USD 0.039.

Enable tiered storage of hot data and cold data

1. Log on to the ApsaraDB for ClickHouse console.

2. In the top navigation bar, select the region where your cluster is deployed.

3. On the Clusters page, click the Clusters of Community-compatible Edition tab and click the ID of the cluster that you want to manage.

4. In the left-side navigation pane, click Cold and hot layered storage.

5. Click Click to Enable Layered Storage.

6. In the Enable Layered Storage message, click OK.

   Then, the system restarts the cluster. After the state of the cluster changes from Changing Configuration to Running, tiered storage of hot data and cold data is enabled.

Default storage policy used after tiered storage of hot data and cold data is enabled

The following table describes the parameters for the default storage policy that is used after tiered storage of hot data and cold data is enabled.

Add a TTL-based tiered storage policy

Add a TTL clause to the default storage policy. This way, all data generated before the specified period in the TTL clause is automatically migrated to a cold data disk.

TTL <time_column> + INTERVAL <number> TO DISK 'cold_disk'

Example

CREATE TABLE ttl_test_tbl
(
    `f1` String,
    `f2` String,
    `f3` Int64,
    `f4` Float64,
    `date` Date
)
ENGINE = MergeTree()
PARTITION BY date
ORDER BY f1
TTL date + INTERVAL 90 DAY TO DISK 'cold_disk'
SETTINGS storage_policy = 'hot_to_cold';

Modify TTL-based tiered storage of hot data and cold data

You can change the column of the DATE or DATETIME data type and the period in a TTL clause to modify the TTL-based tiered storage policy.

ALTER TABLE <table_name> ON CLUSTER default MODIFY TTL <time_column> + INTERVAL <number> TO DISK 'cold_disk';

Migrate data on hot data disks and cold data disks

• Execute the following statement to migrate data from a hot data disk to a cold data disk:

  ALTER TABLE <table_name> ON CLUSTER default MOVE PARTITION <partition> TO DISK 'cold_disk';

• Execute the following statement to migrate data from a cold data disk to a hot data disk:

  ALTER TABLE <table_name> ON CLUSTER default MOVE PARTITION <partition> TO DISK 'default';

  Note

  partition specifies a partition name. You can view the value in the system.parts system table.

View data on hot data disks and cold data disks

• Execute the following statement to view data on a hot data disk:

  SELECT * FROM system.parts WHERE database = '<db_name>' AND TABLE = '<tbl_name>' AND disk_name ='default' AND active = 1;

• Execute the following statement to view data on a cold data disk:

  SELECT * FROM system.parts WHERE database = '<db_name>' AND TABLE = '<tbl_name>' AND disk_name ='cold_disk' AND active = 1;

View the disk space

SELECT * FROM system.disks;

┌─name─────┬─────────path───────────────────────────┬─free_space───────────┬─total_space ─────────┬─used_space───┬─keep_free_space───┬──type──┐
│ cold_disk│ /clickhouse/data/data/disks/cold_disk/ │ 18446744073709551615 │ 18446744073709551615 │ 115312080698 │   0               │   OSS  │
│ default  │ /clickhouse/data/data/                 │ 201663209472         │ 207083249664         │ 5420040192   │   0               │  local │ 
└──────────┴────────────────────────────────────────┴──────────────────────┴──────────────────────┴──────────────┴───────────────────┴────────┘