通过配置低频存储、归档存储降低日志存储成本。-Simple Log Service(SLS)-阿里云帮助中心

Simple Log Service (SLS) provides intelligent tiered storage that lets you store data in hot, infrequent access, or archive storage tiers based on your needs. This reduces long-term storage costs without affecting log capabilities such as query, analysis, visualization, alerting, shipping, and transformation.

Intelligent tiered storage

Storage types

Type	Description
hot storage	Hot storage is a scalable, highly available solution for frequently accessed data. Supports real-time data access with high-performance log query and analysis, ideal for high-frequency data analysis.
infrequent access storage	Infrequent access storage (formerly cold storage) is a cost-effective storage type for long-term data retention. It supports all log capabilities, including query, analysis, visualization, alerting, shipping, and transformation. Ideal for less frequent queries and analysis, such as problem investigation.
archive storage	Archive storage extends hot and infrequent access storage with a lower-cost, long-term solution that remains queryable and analyzable. Designed for long-term data retention for auditing purposes.

Storage periods

Important

The total data retention period is the sum of the retention periods for hot storage, infrequent access storage, and archive storage.
A data retention period of 3,650 days indicates permanent storage. Data continues to be stored after 3,650 days and is billed as archive storage. To learn how to view the data retention period, see Set the data retention period and delete logs.

Data retention period of less than 3,650 days

Scenario 1: The total data retention period is 90 days. Data is stored in the hot storage tier for 30 days and then automatically transitions to the archive storage tier, where it is stored for 60 days before being automatically deleted.

Scenario 2: The total data retention period is 97 days. Data is stored in the hot storage tier for 7 days and then automatically transitions to the infrequent access storage tier. After 30 days in the infrequent access storage tier, it transitions to the archive storage tier, where it remains for 60 days before being automatically deleted.
Data retention period of 3,650 days or more

Scenario 3: The total data retention period is 3,650 days, and the hot storage retention period is 30 days. After 30 days, all data transitions to the archive storage tier. The retention period for archive storage must be set to 3,650 - 30 = 3,620 days. Because this setting enables permanent storage, data remains in the archive storage tier even after 3,620 days.

Scenario 4: The total data retention period is 3,650 days, with a hot storage period of 7 days and an infrequent access storage period of 30 days. The retention period for archive storage must be set to 3,650 - 7 - 30 = 3,613 days. Because this setting enables permanent storage, data remains in the archive storage tier even after 3,613 days.

Comparison of storage types

Item		hot storage	infrequent access storage	archive storage
Use cases		High-frequency, high-performance query and analysis	Problem investigation and analysis	Data audit
Cost		CNY 0.0115 per GB/day	CNY 0.005 per GB/day	CNY 0.0017 per GB/day
Performance	Latency (billions of records)	Tens to hundreds of milliseconds	Hundreds of milliseconds to seconds	Minutes
Performance	Concurrency (Project-level)	Concurrent queries: 100 Concurrent analyses: 2	Concurrent queries: 10 Concurrent analyses: 2	Concurrent queries: 1 Concurrent analyses: 1
Benefits		Fast, high-concurrency query and analysis	Cost-effective for infrequent query and analysis	Lowest cost for occasional query and analysis
Benefits		Data automatically transitions between storage tiers based on lifecycle management rules: Simple configuration with no scripting or manual data migration required. Data in infrequent access and archive storage can be accessed in real time, just like hot storage data. No manual retrieval, application changes, or retrieval fees required.
Storage duration limits	pay-by-feature billing	Data must be stored in hot storage for at least 7 days before transitioning to infrequent access storage. Data must be stored in hot storage for at least 30 days before transitioning to archive storage.	Data must be stored in infrequent access storage for at least 30 days before transitioning to archive storage.	The minimum storage duration is 60 days.
Storage duration limits	pay-by-ingested-data	A 30-day free hot storage quota is included.		The minimum storage duration is 60 days.

Storage tier transition process

Data can transition between the three storage tiers. The following table describes the transition rules.

No.	Description
1	Data in hot storage must be stored for at least 7 days before transitioning to infrequent access storage.
2	Data in hot storage must be stored for at least 30 days before transitioning to archive storage.
3	Data in infrequent access storage must be stored for at least 30 days before transitioning to archive storage.
4	You can transition data from infrequent access storage back to hot storage by modifying the hot storage retention period.
5	You can transition data from archive storage back to infrequent access storage by modifying the infrequent access storage retention period.
6	You can transition data from archive storage back to hot storage by modifying the hot storage retention period.

Manage storage tiers

Console

Log on to the Simple Log Service console.
In the Projects section, click the one you want.
On the Log Storage > Logstores tab, find the target Logstore, click the icon, and then click Modify.

In the Logstore Attributes panel, click Modify. Turn on Intelligent Tiered Storage, configure the parameters, and then click Save.

The following table describes the intelligent tiered storage parameters. For more information about other parameters, see Create a basic Logstore.

Parameter	Description
Data Retention Period	You can set the Data Retention Period to Specified Days or Permanent Storage. The total data retention period is the sum of the retention periods for the hot, infrequent access, and archive storage tiers. Important If you shorten the data retention period, the change takes effect within one minute. For example, if you change the retention period from 5 days to 1 day, SLS deletes the data from the previous 4 days. The actual data deletion might be delayed by up to 7 days, but this delayed data will not incur charges or count toward your storage capacity.
Intelligent Tiered Storage	Turn on Intelligent Tiered Storage to enable the feature.
Storage Policy	Configure the retention period for each storage tier.
Hot Data Retention	How long data is stored in the hot storage tier of the Logstore. For transitions from hot storage to infrequent access storage, the valid range is 7 to 3,650 days. For transitions from hot storage to archive storage, the valid range is 30 to 3,650 days. Important After you modify the Hot Data Retention period, the change takes effect within one minute. For example, if you change the hot storage retention period from 30 days to 40 days, SLS starts to transition data from infrequent access or archive storage back to hot storage within one minute. The value of Hot Data Retention must be less than the value of Data Retention Period.
After the data in the IA storage tier is retained for more than	How long data is stored in the infrequent access storage tier of the Logstore. For transitions from infrequent access storage to archive storage, the valid range is 30 to 3,650 days. Important After you modify the Infrequent Access Data Retention period, the change takes effect within one minute. For example, if you change the infrequent access storage retention period from 30 days to 40 days, SLS starts to transition data from archive storage back to infrequent access storage within one minute. The value of Infrequent Access Data Retention must be less than the value of Data Retention Period.
After the data in the Archive storage tier is retained for more than	How long data is stored in the archive storage tier of the Logstore. Valid values: 60 to 3650 days. When data exceeds the configured After the data in the Archive storage tier is retained for more than period, it is automatically deleted. Important The value of Archive Data Retention must be less than the value of Data Retention Period.

API

When you create a Logstore, call the CreateLogStore operation and pass the ttl (total data retention period), hot_ttl (hot storage retention period), and infrequentAccessTTL (infrequent access storage retention period) parameters to configure the tiered storage policy.
For an existing Logstore, call the UpdateLogStore operation to update the ttl, hot_ttl, and infrequentAccessTTL parameters to dynamically adjust the tiered storage policy.

Manage intelligent tiered storage