[英]Azure Synapse Analytics monitor rowgroup skipping
For reference i'm using a CCI table (Hash) I'm looking for a solution to monitor if, for a query with a WHERE predicate on my Hash, Synapse is well skipping the invalid rowgroups.作为参考,我正在使用 CCI 表(哈希)我正在寻找一种解决方案来监视,对于在我的 Hash 上使用 WHERE 谓词的查询,Synapse 是否很好地跳过了无效的行组。
For a Hash-distributed table Synapse will eliminate Distributions, not rowgroups.对于散列分布表 Synapse 将消除分布,而不是行组。 It's the partition scheme and CCI order that control row group skipping.
控制行组跳过的是分区方案和 CCI 顺序。 And that should be evident in the execution plan .
这在执行计划中应该很明显。
For a CCI, each table has multiple (60) Distributions.对于 CCI,每个表都有多个 (60) 分布。 Each Distribution has one or more Partitions.
每个分布都有一个或多个分区。 Each Partition has one or more Row Groups.
每个分区都有一个或多个行组。 Each Row Group has up to 1 million rows.
每个行组最多有 100 万行。
For Heaps and Clustered Index tables it's the same, except each Partition has one Heap or Clustered Index.对于堆和聚集索引表,它是相同的,除了每个分区都有一个堆或聚集索引。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.