Session One: Moving Beyond Tables
"Google doesn't store your GMail in a MySQL table. Your Amazon shopping cart doesn't act_as_list. Orbitz doesn't run a SELECT to find the best flights. But when choosing technology, nearly every new web startup will reach for a relational database. Why? Can we do better?"
Avi Bryant has a beard. He's from Vancouver. He's not from Silicon Valley. His words, not mine.
And we begin.
Common Themes of large data sets.
- No Joins, no grouping, no aggregation, restricted sorting, restricted CPU and a dynamic schema.
Being forced into restricted query calls (with forced sorting), requires doing your own sorting (if you require it) at the application layer.
I don't really know where the Avi is going with this conversation yet, he's starting pretty high up the path.
