Indexes are one of the most important pieces in performance tuning – they can help queries seek to find a small number of records, efficiently scan for sets, use batch mode processing for large scale aggregates / join. Or, sadly, they can be a complete MESS! The basic problem is that EVERY SINGLE query *can* be indexed. But, not every query *should* be indexed. I’ll start with a quick review of row-based and column-based indexes but I’ll primarily focus on strategies that will help your server perform! We’ll start with the ever-important “query tuning” aspect to indexing but also cover the many considerations overlooked before applying those indexes in production (what I call “server tuning with indexes”). This will be a fast-paced session with lots of demos, get your caffeine first!