Thanks to Paul Randal and his recently published blog post: “The Curious Case of… tracking page compression success rates“, I finally discovered what was the purpose of the column page_compression_success_count from the system function sys.dm_db_index_operational_stats, and how to use it.
Read More »Detect Low PAGE Compression Success Rates in all Databases
You might have seen this blog post by Paul White: New Metadata-Only Column Changes in SQL Server 2016. It talks about how, when a table is compressed, you can change (i.e. increase) a column’s data type without having to create a new table and migrate the data into it, and without having to refactor all of the underlying data. But, not everything is rainbows and unicorns…
Read More »The new “metadata-only column changes” feature in SQL Server 2016 is useless!
Data compression in SQL Server can definitely go a long way in saving storage space, as well as improve performance. It can especially be significant in very large databases. But in certain scenarios it can actually cause more harm than good, and it’s important to identify these scenarios.
I wrote this script which performs compression savings estimation checks for an ENTIRE database. Yeah there are plenty such scripts out there already, but there’s something in particular that they all seem to be missing, and that is the ability to actually generate the relevant remediation commands for you. After all, ain’t nobody got time for writing all those “ALTER TABLE/INDEX” commands themselves!
This is why I created the T-SQL script below. Check out the “messages” tab once the script completes execution, to find the resulting remediation commands. In addition, you will also get a query result with full details.
The script also performs various checks to determine whether PAGE or ROW compression type would be more beneficial (based on stuff such as range scans vs. updates percentage, difference in compression savings, etc.). It has several parameters at the top to control various thresholds and behaviors, so that you could fine tune it for your specific database.
Read More »The Ultimate Compression Savings Estimation Script for an Entire Database