Sign InTry Free

GC Configuration

The GC (Garbage Collection) configuration and operational status are recorded in the mysql.tidb system table. You can use SQL statements to query or modify them:

select VARIABLE_NAME, VARIABLE_VALUE from mysql.tidb where VARIABLE_NAME like "tikv_gc%";
+--------------------------+----------------------------------------------------------------------------------------------------+ | VARIABLE_NAME | VARIABLE_VALUE | +--------------------------+----------------------------------------------------------------------------------------------------+ | tikv_gc_leader_uuid | 5afd54a0ea40005 | | tikv_gc_leader_desc | host:tidb-cluster-tidb-0, pid:215, start at 2019-07-15 11:09:14.029668932 +0000 UTC m=+0.463731223 | | tikv_gc_leader_lease | 20190715-12:12:14 +0000 | | tikv_gc_enable | true | | tikv_gc_run_interval | 10m0s | | tikv_gc_life_time | 10m0s | | tikv_gc_last_run_time | 20190715-12:09:14 +0000 | | tikv_gc_safe_point | 20190715-11:59:14 +0000 | | tikv_gc_auto_concurrency | true | | tikv_gc_mode | distributed | +--------------------------+----------------------------------------------------------------------------------------------------+ 13 rows in set (0.00 sec)

For example, the following statement makes GC keep history data for the most recent 24 hours:

update mysql.tidb set VARIABLE_VALUE="24h" where VARIABLE_NAME="tikv_gc_life_time";

tikv_gc_enable

  • Enables or disables GC
  • Default: true

tikv_gc_run_interval

  • Specifies the GC interval, in the format of Go Duration, for example, "1h30m", and "15m"
  • Default: "10m0s"

tikv_gc_life_time

  • The time limit during which data is retained for each GC, in the format of Go Duration. When a GC happens, the current time minus this value is the safe point.

  • Default: "10m0s"

tikv_gc_mode

  • Specifies the GC mode. Possible values are:

    • "distributed" (default): Distributed GC mode. In the Do GC step, the GC leader on the TiDB side uploads the safe point to PD. Each TiKV node obtains the safe point respectively and performs GC on all leader Regions on the current node. This mode is is supported from TiDB 3.0.

    • "central": Central GC mode. In the Do GC step, the GC leader sends GC requests to all Regions. This mode is adopted by TiDB 2.1 or earlier versions.

tikv_gc_auto_concurrency

  • Controls whether to let TiDB automatically specify the GC concurrency, or the maximum number of GC threads allowed concurrently.

    When tikv_gc_mode is set to "distributed", GC concurrency works in the Resolve Locks step. When tikv_gc_mode is set to "central", it is applied to both the Resolve Locks and Do GC steps.

    • true(default): Automatically use the number of TiKV nodes in the cluster as the GC concurrency
    • false: Use the value of tikv_gc_concurrency as the GC concurrency

tikv_gc_concurrency

  • Specifies the GC concurrency manually. This parameter works only when you set tikv_gc_auto_concurrency to false.
  • Default: 2

tikv_gc_scan_lock_mode (experimental feature)

This parameter specifies the way of scanning locks in the Resolve Locks step of GC, that is, whether to enable Green GC (experimental feature) or not. In the Resolve Locks step of GC, TiKV needs to scan all locks in the cluster. With Green GC disabled, TiDB scans locks by Regions. Green GC provides the "physical scanning" feature, which means that each TiKV node can bypass the Raft layer to directly scan data. This feature can effectively mitigate the impact of GC wakening up all Regions when the Hibernate Region feature is enabled, thus improving the execution speed in the Resolve Locks step.

  • "legacy" (default): Uses the old way of scanning, that is, disable Green GC.
  • "physical": Uses the physical scanning method, that is, enable Green GC.

Notes on GC process changes

Since TiDB 3.0, some configuration options have changed with support for the distributed GC mode and concurrent Resolve Locks processing. The changes are shown in the following table:

Version/ConfigurationResolve LocksDo GC
2.xSerialConcurrent
3.0
tikv_gc_mode = centered
tikv_gc_auto_concurrency = false
ConcurrentConcurrent
3.0
tikv_gc_mode = centered
tikv_gc_auto_concurrency = true
Auto-concurrentAuto-concurrent
3.0
tikv_gc_mode = distributed
tikv_gc_auto_concurrency = false
ConcurrentDistributed
3.0
tikv_gc_mode = distributed
tikv_gc_auto_concurrency = true
(default)
Auto-concurrentDistributed
  • Serial: requests are sent from TiDB Region by Region.
  • Concurrent: requests are sent to each Region concurrently based on the number of threads specified in the tikv_gc_concurrency.
  • Auto-concurrent: requests are sent to each Region concurrently with the number of TiKV nodes as concurrency value.
  • Distributed: no need for TiDB to send requests to TiKV to trigger GC because each TiKV handles GC on its own.

In addition, if Green GC (experimental feature) is enabled, that is, setting the value of tikv_gc_scan_lock_mode to physical, the processing of Resolve Lock is not affected by the concurrency configuration above.

GC I/O limit

TiKV supports the GC I/O limit. You can configure gc.max-write-bytes-per-sec to limit writes of a GC worker per second, and thus to reduce the impact on normal requests.

0 indicates disabling this feature.

You can dynamically modify this configuration using tikv-ctl:

tikv-ctl --host=ip:port modify-tikv-config -n gc.max-write-bytes-per-sec -v 10MB

Was this page helpful?

Download PDFRequest docs changesAsk questions on Discord
Playground
New
One-stop & interactive experience of TiDB's capabilities WITHOUT registration.
Products
TiDB
TiDB Dedicated
TiDB Serverless
Pricing
Get Demo
Get Started
© 2024 PingCAP. All Rights Reserved.
Privacy Policy.