SQLBits 2012

SQL Server Indexes : The forgotten Fill-Factor

The fill-factor index option has a huge impact on the performance of your DB. By using a different approach for specific use cases this session will give you the tools to find the most optimal fill-factor for your tables.

While we all know the fill factor index option when creating or rebuilding an index, few know about the huge impact a fill factor has on the performance of your machine. Most DBA's treat the fill-factor as a static server setting and "forget about it".

“What number should I set the fill-factor to?” is a very challenging question that most DBA’s should bother about. High fill-factors have high data densities, while low fill-factors can cope better with nonsequential IO. The challenge is to find a statistical model that will minimize fragmentation, optimize insert performance while reducing disk usage overhead and all this without stressing the maintenance window.

By using a different approach for specific use cases this session will give you the tools to find the most optimal fill-factor for your tables.

Speaker

Karel Coenye a senior SQL Server trainer and Consultant at Kohera, has been working with SQL Server in different roles since 1999. Currently working on different database projects as SQL architect while coaching and training other DBAs and developers. Experienced in both full development-environments as in production environments with large server with multiple large databases specialized in tuning, tweaking and monitoring. His IT passion drives him towards the physical and virtual bottlenecks while seeking ways to push systems towards their limits.

Karel Coenye's Sessions

SQL Server Indexes : The forgotten Fill-Factor