NLS and data "temperatures"

[henry_jones]

When I look at NLS presentations, they often classify data access using hot, warm, and cold. How can this be determined?

[Former Member]

Just have a look in your statistics and find out how often your cube/DSO to archive is used.

If it's used daily then it'S hot.

If it's used frequently (weekly, monthly) then it's warm

if it'S used seldomly or not at all then it'S cold.

But the exact defintion of data age and how long to preserver the data depends on your business and sometimes  legal regulations.

Regards,

Jürgen

[henry_jones]

Thanks for the reply, but I'm trying to determine how often the time periods within a cube/DSO are being accessed.

For example, how often is time period 2003 accessed from cube XYZ.

Thanks for your input.

[Former Member]

Hi,,you need query selections for this.

May you can use a query from Query details from Technical content.

Copy the query to your own namespace and change it to your needs.

With aggregates there was a simple way. Just create an aggregate for year or periods and the aggregate counter told you how often the aggregate was used.

MAy you can also use some cache information.

But that's quite tedious work to find out how often a period is accessed.

Regards,

Jürgen