The TRUNCATE TABLE statement is a DDL (Data Definition Language) operation that removes all rows from a table but leaves the table metadata intact, including the table schema, description, and labels. Since TRUNCATE is a metadata operation it does not incur a charge.
Hi Mahesh, Excellent Work you are doing, I have one small query, can we implement some rules to preempt the BQ queries that would run an excessive workload (e.g. query processing more than 50Gig, or similar spikes)? Please share your comments on this.
Great Tips Mahesh, I think we can also do partitioning if we know the user is firing a query on particular column like region frequently and as I know if single partitioning goes beyond 1 Gb then we can go for clustering so less data can be scanned and it can save cost
The TRUNCATE TABLE statement is a DDL (Data Definition Language) operation that removes all rows from a table but leaves the table metadata intact, including the table schema, description, and labels. Since TRUNCATE is a metadata operation it does not incur a charge.
How did you write the ddl, which language,is it in SQL?
@@adesojialu1051 Sorry, I did not understand the question
Learning from mistakes and sharing with everyone is important to save the cost, It is very informative!
wow you are always the king
Thanks
This has been helful!
Great.. Do extend this for remaining services as well!!
Sure, Shanmuk
Great video Mahesh! Thanks.
Nice tips ,how to analyse reports here in bq for performance part.
Hi Mahesh,
I think we can consider to use the partitioning feature? to have cost optimization.
True
Hi Mahesh, Excellent Work you are doing, I have one small query, can we implement some rules to preempt the BQ queries that would run an excessive workload (e.g. query processing more than 50Gig, or similar spikes)? Please share your comments on this.
very nice tips
But truncate can be done by running the query in BQ which incurs certain charges right?
No it will not incur cost
Great Tips Mahesh, I think we can also do partitioning if we know the user is firing a query on particular column like region frequently and as I know if single partitioning goes beyond 1 Gb then we can go for clustering so less data can be scanned and it can save cost
True
Hi Mahesh....Do you provide GCP and Devops training too
is it the network cost that is optimized if we create in the same region?
Network cost - Egress will be reduced
thanks sir
Thanks Gaurav