Skip to content

reference/performance: correct replica selection policy of follower read #2289

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
yikeke merged 3 commits into from
Feb 24, 2020
Merged

reference/performance: correct replica selection policy of follower read #2289

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
25f51dd
reference/performance: correct replica selection policy of follower read
Feb 24, 2020
b04b60a
Merge 'master' into follower-read
Feb 24, 2020
d14376d
Merge branch 'master' into follower-read
sre-bot Feb 24, 2020
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion reference/performance/follower-read.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -47,4 +47,4 @@ TiKV follower 节点处理读取请求时，首先使用 Raft `ReadIndex` 协议

### Follower 副本选择策略

由于 TiKV 的 Follower Read 可以保证线性一致性，不会破坏 TiDB 的 Snapshot Isolation 事务隔离级别，因此 TiDB 选择 follower 的策略可以采用 round robin 的方式。虽然 TiKV 可以选择任意的 follower 处理任意读取请求，但考虑到多个 follower 间复制速度不同，如果负载均衡的粒度过细，可能会导致明显的 latency 波动。目前，Follower Read 负载均衡策略粒度是连接级别的，对于一个 TiDB 的客户端连接在某个具体的 Region 上会固定使用同一个 follower，只有在选中的 follower 发生故障或者因调度策略发生调整的情况下才会进行切换。
由于 TiKV 的 Follower Read 可以保证线性一致性，不会破坏 TiDB 的 Snapshot Isolation 事务隔离级别，因此 TiDB 选择 follower 的策略可以采用 round robin 的方式。目前，Follower Read 负载均衡策略粒度是事务级别的，对于一个 TiDB 的事务在某个具体的 Region 上会固定使用同一个 follower，只有在选中的 follower 发生故障或者因调度策略发生调整的情况下才会进行切换。