mysql - Understanding Distribution -
mysql - Understanding Distribution -
i have couple of questions.
i know if need worry distribution in netezza while using select statements(not creating tables). trying create dataset in sas connecting netezza , selecting view has couple of joins. wondering how impact performance of netezza if creating table straight in sas.
i creating table joining 2 tables on customer_id. however, output dataset not consist of customer_id column. can distribute table on customer_id?
thanks.
for first question, typically don't need worry distribution if aren't creating table. help understand distribution methods tables selecting from, it's not requirement. having distribution method supports particular joins doing can help performance during select (e.g. if bring together columns superset of distribution columns you'll co-located joins), if target of output sas, there's no effect on write of dataset sas.
for sec question, table distributed either on column, or columns, in table itself, or via random (aka round robin) distribution method. in case, if storing info set in table on netezza, not distribute info on customer_id column not included in info set.
mysql sql sas netezza
Comments
Post a Comment