From 7167ce455bc689fe7d30fb415b8bbebcdbbece6b Mon Sep 17 00:00:00 2001 From: qiancai Date: Fri, 22 Dec 2023 11:53:31 +0800 Subject: [PATCH] Update release-6.1.0.md --- releases/release-6.1.0.md | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-) diff --git a/releases/release-6.1.0.md b/releases/release-6.1.0.md index 4b555be897299..c230f0a289d5e 100644 --- a/releases/release-6.1.0.md +++ b/releases/release-6.1.0.md @@ -59,9 +59,15 @@ In 6.1.0, the key new features or improvements are as follows: ### Performance -* Support customized Region size (experimental) +* Support customized Region size - Setting Regions to a larger size can effectively reduce the number of Regions, make Regions easier to manage, and improve the cluster performance and stability. This feature introduces the concept of bucket, which is a smaller range within a Region. Using buckets as the query unit can optimize concurrent query performance when Regions are set to a larger size. Using buckets as the query unit can also dynamically adjust the sizes of hot Regions to ensure the scheduling efficiency and load balance. This feature is currently experimental. It is not recommended to use it in production environments. + Starting from v6.1.0, you can configure [`coprocessor.region-split-size`](/tikv-configuration-file.md#region-split-size) to set Regions to a larger size. This can effectively reduce the number of Regions, make Regions easier to manage, and improve the cluster performance and stability. + + [User document](/tune-region-performance.md), [#11515](https://github.com/tikv/tikv/issues/11515) + +* Support using buckets to increase concurrency (experimental) + + To help you further improve the query concurrency after setting Regions to a larger size, TiDB introduces the concept of bucket, which is a smaller range within a Region. Using buckets as the query unit can optimize concurrent query performance when Regions are set to a larger size. Using buckets as the query unit can also dynamically adjust the sizes of hot Regions to ensure the scheduling efficiency and load balance. This feature is currently experimental. It is not recommended to use it in production environments. [User document](/tune-region-performance.md), [#11515](https://github.com/tikv/tikv/issues/11515)