Hi,
Added an index maintenance job (Hallengren) to a database (SQL Server 2016) with a few large tables.
Largest table has around 500.000.000 rows with a size of around 250 GB (3 indexes totally 125 GB so total data also around 125 GB).
The script will do a reorg if fragmentation goes over 5 %. This happens approx. every 2-3 day if running maintenance once per day (night).
So far, all good. My problem is that the reorg is running for a very long time. For example, the heaviest index (clustered index and also pk) is running reorg for around 4 hours 30 min even do the index was only fragmented to 5,33%.
My server is a quite thin one. Virtual with 16 GB RAM and 2 CPU cores (quite new Xeon).
Read somewhere that reorg is single-threaded.
Can I assume the reorganize time to decrease if I boost the server with more cores?
Any other suggestion?
Thanks!