From 04eb51495a60da504849e609625a0f71bdaba4f6 Mon Sep 17 00:00:00 2001 From: RunningDB <144221862+RunningDB@users.noreply.github.com> Date: Wed, 25 Dec 2024 09:18:19 +0800 Subject: [PATCH] [docs] Fix typo in website/docs/table-design/data-distribution/bucketing.md (#261) --- website/docs/table-design/data-distribution/bucketing.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/website/docs/table-design/data-distribution/bucketing.md b/website/docs/table-design/data-distribution/bucketing.md index 23b40ae1..17274d96 100644 --- a/website/docs/table-design/data-distribution/bucketing.md +++ b/website/docs/table-design/data-distribution/bucketing.md @@ -9,7 +9,7 @@ and distributes the data to multiple hosts and services. When creating a Fluss table, you can specify the number of buckets by setting `'bucket.num' = ''` property for the table, see more details in [DDL](/docs/engine-flink/ddl.md). Currently, Fluss supports 3 bucketing strategies: **Hash Bucketing**, **Sticky Bucketing** and **Round-Robin Bucketing**. -Primary-Key Tables only allows to use **Hash Bucketing**. Log Tables use **Sticky Bucketing** by default but can use other two bucketing strategies. +Primary-Key Tables only allow to use **Hash Bucketing**. Log Tables use **Sticky Bucketing** by default but can use other two bucketing strategies. ## Hash Bucketing **Hash Bucketing** is common in OLAP scenarios. @@ -21,7 +21,7 @@ Primary-Key Tables use primary key (excluding partition key) as the bucket key b ## Sticky Bucketing -**Sticky Bucketing** enables larger batches and reduce latency when writing records into Log Tables. After sending a batch, the sticky bucket changes. Over time, the records are spread out evenly among all the buckets. +**Sticky Bucketing** enables larger batches and reduces latency when writing records into Log Tables. After sending a batch, the sticky bucket changes. Over time, the records are spread out evenly among all the buckets. Sticky Bucketing is the default bucketing strategy for Log Tables. This is quite important because Log Tables uses Apache Arrow as the underling data format which is efficient for large batches. **Usage**: setting `'client.writer.bucket.no-key-assigner'='sticky'` property for the table to enable this strategy. PrimaryKey Tables do not support this strategy.