When a GROUP BY hash table can't fit in memory, sometimes the external sort that is triggered is extremely CPU intensive. If we find it impossible or prohibitively difficult to optimize such queries, is there a way to either force these queries to error out, or limit their CPU usage?
Possible to force queries to abort instead of doing large external sort groupings?
With Vertica 7, you can limit queries on a resource pool to a certain number of cores. You can also set a RUNTIMECAP, which will cause the query to abort if it runs too long (works pre 7.0). There isn't currently a way to control the specific case of group by hash spill.
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
