geert.de.vylder (3/6/2013)
ChrisM@Work (3/6/2013)
I meant in terms of potentially reducing the rowcount.Let's call a set from this table containing the same day, all of the intervals, and the same Event_Channel, a frame. One frame contains 96 rows, one row per interval. There are 9 event channels so there are 864 rows per date. Multiply 864 by the number of dates in the tickets table - 1366 - to yield the number of rows in the matrix table.
One possibility is to remove frames which don't have any corresponding rows in the tickets table.
Sorry,
I misunderstood you.
Then the best would be to work with a variable of the datetime type and set this variable to a fix date (for example 2013-03-01 YYYY-MM-DD format) or to set it to yesterdays date. And then retreive the frame for all 9 event channels.
Would that be better?
Greetz,
Geert
I've no idea, Geert. Folks here on SSC will help you with coding decisions; this is more of a business decision. A date filter seems sensible to me but is it appropriate for the consumers of the data? How many days' worth of data are other folks expecting to see?
For fast, accurate and documented assistance in answering your questions, please read this article.
Understanding and using APPLY, (I) and (II) Paul White
Hidden RBAR: Triangular Joins / The "Numbers" or "Tally" Table: What it is and how it replaces a loop Jeff Moden