[GitHub] [flink-table-store] LadyForest commented on a change in pull request #72: [FLINK-26899] Introduce write/query table document for table store

[GitBox <gi...@apache.org>]

LadyForest commented on a change in pull request #72:
URL: https://github.com/apache/flink-table-store/pull/72#discussion_r841147565



##########
File path: docs/content/docs/development/create-table.md
##########
@@ -151,3 +151,85 @@ Creating a table will create the corresponding physical storage:
 - If `log.system` is configured as Kafka, a Topic named
   "${catalog_name}.${database_name}.${table_name}" will be created
   automatically when the table is created.
+
+## Distribution
+
+The data distribution of Table Store consists of three concepts:
+Partition, Bucket, and Primary Key.
+
+```sql
+CREATE TABLE MyTable (
+  user_id BIGINT,
+  item_id BIGINT,
+  behavior STRING,
+  dt STRING,
+  PRIMARY KEY (dt, user_id) NOT ENFORCED
+) PARTITIONED BY (dt) WITH (
+  'bucket' = '4'
+);
+```
+
+For example, the `MyTable` table above has its data distribution
+in the following order:
+- Partition: isolating different data based on partition fields.
+- Bucket: Within a single partition, distributed into 4 different
+  buckets based on the hash value of the primary key.
+- Primary key: Within a single bucket, sorted by primary key to
+  build LSM structure.
+
+## Partition
+
+Table Store adopts the same partitioning concept as Apache Hive to
+separate data, and thus various operations can be managed by partition
+as a management unit.
+
+Partitioned filtering is the most effective way to improve performance,
+your query statements should contain partition filtering conditions
+as much as possible.
+
+## Bucket
+
+The record is hashed into different buckets according to the
+primary key or the whole row (without primary key).
+
+The number of buckets is very important as it determines the
+worst-case maximum processing parallelism. But it should not be
+too big, otherwise, the system will create a lot of small files.
+
+In general, the desired file size is 128 MB, the recommended data
+to be kept on disk in each sub-bucket is about 1 GB.
+
+## Primary Key
+
+The primary key is unique and is indexed.

Review comment:
       Nit: The primary key is unique and indexed.




-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: issues-unsubscribe@flink.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org