- Mais recentes
- Mais votos
- Mais comentários
Hi,
On Postgresql, the pg_stat_statements module provides a means for tracking planning and execution statistics of all SQL statements executed by a server.
See https://www.postgresql.org/docs/current/pgstatstatements.html
To view SQL digest statistics, the pg_stat_statements library must be loaded.
For Aurora PostgreSQL DB clusters that are compatible with PostgreSQL 10,
this library is loaded by default. For Aurora PostgreSQL DB clusters that are
compatible with PostgreSQL 9.6, you enable this library manually. To enable
it manually, add pg_stat_statements to shared_preload_libraries in the DB
parameter group associated with the DB instance.
Additionally, you should turn on track_io_timing on your Aurora instance: see section "Aurora PostgreSQL cluster-level parameters" of https://docs.aws.amazon.com/AmazonRDS/latest/AuroraUserGuide/AuroraPostgreSQL.Reference.ParameterGroups.html
See https://www.postgresql.org/docs/current/runtime-config-statistics.html for details on this parameter.
Combining all the above will allow you to track the requests consuming most I/Os and to optimize them.
Best,
DIdier
To add on to above answer, you may use below query to get the list of top 10 SQL queries having high I/O activity:
SELECT userid::regrole, dbid, query
FROM pg_stat_statements
ORDER BY blk_read_time + blk_write_time DESC
LIMIT 10;
Note: Make sure that you have pg_stat_statements extension installed and have track_io_timing enabled (i.e. track_io_timing=1)
@aws_abhi Love it, thanks. How does "blk_*_time" correlate with the IOPS?
Conteúdo relevante
- AWS OFICIALAtualizada há 2 anos
Thank you @Didier_Durand