Great video sir. I have one doubt. How much time Snowflake would take to implement manual/custom clustering since it involves reshuffling of data ? Is it created instantly? Any way to get the status of manually clustered table?
Hi, Clustering works a little differently in Snowflake vs Traditional databases, Snowflake uses micro-partition (find out more here docs.snowflake.com/en/user-guide/tables-micro-partitions). This means that Snowflake automatically splits data into micro partitions when it is loaded. As data grows and these original clusters become non-optimal you can designate one or more clustering keys on a table with: CREATE TABLE ... CLUSTER BY ( [ , ... ] ) Depending on why you're looking to re-cluster, there are alternative options that may be suitable based on their use case, such as Query Acceleration Service, optimising queries, and various other methods. (Also note that Snowflake only recommends clustering on large tables-at least TB scale. If your table size is a few GB, then improving the SQL query will contribute more than clustering).
Even after 2 years, this single video is much better than the time waste multi part series of behind the cape.
Very informative. Thank You so much @Pat Lucas.
Glad you found it was helpful!
Great video sir.
I have one doubt. How much time Snowflake would take to implement manual/custom clustering since it involves reshuffling of data ?
Is it created instantly?
Any way to get the status of manually clustered table?
Hi, Clustering works a little differently in Snowflake vs Traditional databases, Snowflake uses micro-partition (find out more here docs.snowflake.com/en/user-guide/tables-micro-partitions). This means that Snowflake automatically splits data into micro partitions when it is loaded.
As data grows and these original clusters become non-optimal you can designate one or more clustering keys on a table with:
CREATE TABLE ... CLUSTER BY ( [ , ... ] )
Depending on why you're looking to re-cluster, there are alternative options that may be suitable based on their use case, such as Query Acceleration Service, optimising queries, and various other methods.
(Also note that Snowflake only recommends clustering on large tables-at least TB scale. If your table size is a few GB, then improving the SQL query will contribute more than clustering).