We have some tables which are created in temporal model. For that I believe TOPK projections could be helpful. However refreshing and creating a fresh new topk projection takes lot of time especially table with 800K + records. Is there any best practices for creating and managing TOPK projections to improve the performance. Additionally, sometimes when using TOPK projections, the query hits the MaxOptMemMB although we set the size this to 400MB.
Question
Is there any better way to improve performance of the TOPK-Projections?
+1Is the table partitioned? That might improve the performance, and allow it some "breathing room" for the calculation. The order by clause of the projection is also gonna have a pretty significant impact, as well as the "depth" of the top-k. So, if you're drilling down a bunch of columns into the top-k, it's gonna have to do a lot of work as well. Can you provide some more specifics about what you're creating and what it looks like?
Sign up
Already have an account? Login
Welcome to the Rocket Forum!
Please log in or register:
Employee Login | Registration Member Login | RegistrationEnter your E-mail address. We'll send you an e-mail with instructions to reset your password.
77 4th Avenue
Waltham, MA 02451 USA
