top of page

Identify slow running queries in snowflake

select distinct count(query_id) as total_query

--,QUERY_TEXT

--,DATABASE_NAME

--,SCHEMA_NAME

--,QUERY_TYPE

--,USER_NAME

--,ROLE_NAME

--,WAREHOUSE_NAME

--,WAREHOUSE_SIZE

,date_part(day, start_time) as "Day"

--,round(bytes_scanned/1024/1024/1000) as GB_scanned

,SUM(total_elapsed_time) / 60000/60 as TOTAL_elapsed_Hours

--,(partitions_scanned / nullif(partitions_total,0)) * 100 as pct_table_scan

--, percent_scanned_from_cache * 100 as pct_from_cache

--,bytes_spilled_to_local_storage as spill_to_local

--,bytes_spilled_to_remote_storage as spill_to_remote

from snowflake.account_usage.query_history

where --(bytes_spilled_to_local_storage > 1024 * 1024 or

-- bytes_spilled_to_remote_storage > 1024 * 1024 or

-- percentage_scanned_from_cache < 0.1)

--and elapsed_Minutes > 120

--and bytes_scanned > 1024 * 1024

WAREHOUSE_NAME like 'WAREHOUSENAME%'

and START_TIME >= current_date()-31

--group by START_TIME desc;

group by "Day"

order by "Day";

10 views0 comments

Recent Posts

See All

留言


bottom of page