All geek questions in one place

Perform aggregate functions on large, frequently changing datasets

I need to extract some management information (MI) from data that is updated in nightly batches. I will use aggregated functions to generate MI from tables of hundreds of thousands and possibly millions of rows. Information will be displayed on the web page.
A critical factor here is the efficiency of processing SQL Server aggregate functions.
I came across two options for generating data:

  • Record stored procs / views to generate information from raw data that is called every time someone accesses the page
  • Create tables that are updated daily and act like a cache for MI

What is the best approach?

+3
source share
2 answers

Download the values ​​during the night load if the data does not change during the day. This will make the search much faster. If necessary, I am a big fan of pivot tables. In your case, they are necessary!

, , , , SQL Server, Analysis Services. , , , , , , . 10 , , SSAS.

+4

, Analysis Services . .

, , NO EXPAND, .

SQL 2005 Indexed Views

SQL 2008

+3

Source: https://habr.com/ru/post/1715985/

More articles:


All Articles