none
Scale up detection RRS feed

  • Question

  • Hi, I know that SQL Azure provides automatic scale up on bigger nodes when your DB activity grows beyond a give threshold. How can we know when it is going to happen, and at a give time, how do we know on which kind of node our DB sits ? And is there a way to know that we come close to the throttling threshold before throttling errors occur ?

    These days I'm developping (and load testing using VM roles running VS Load Test agents) a solution based on a dynamically scaling Sharded SQL DB, and there seems to be no way to know when we need to provision new shards before we get throttling errors... (and well, for our business, if we get throttling errors, it's already to late).

    Can you provide guidelines on how to proactively detect needs of more DB shards before it's too late ?

    Friday, January 28, 2011 9:48 AM

All replies