I have a db table with following dimensions having a sample( repeats every 5 mins) data of consumption of license feature hosted on a server.
Time : <date> <hr:min> For ex : 17-May-2013 10:30, 17-May-2013 10:35
License Server Name :
License Feature Name :
User Name : User who consume the license feature on the server.
Site : Physical location of user
Group : Department of user
Cost Center : Cost center of user
LIC_USAGE : number of license feature consumed at a given sample
Total Licenses : Total Number of hosted license feature on the server
Site=>Group=>costcenter are hirarchical dimensions i.e cost center is under groups and group is under site.
Site, group, costcenter are frequently changing dimensions, so i can't use a group of tableau to group them.
I want to calculate peak license feature usage for site, group, costcenter on the server. If i do MAX(LIC_USAGE) it gives me peak usage by user only because user is the minimum level of dimensional hirarchy.
I tried to build a new table with group by site, group, cost center excluding user but that gives me peak usage by
cost center.
Could you please suggest how to calculate it
You need to outline your hierarchy in the viz and use a window max of whatever the relevant aggregation is.
In this example I used a hierarchy of Department>Category>Item of Superstore Sales data
Yours would be the hierarchy you described above
my calc is window_max(sum([Sales]))
compute using pane down.
(Your compute using might be different since I think you have one more level to your hierarchy)
On the second sheet I added an index() filter of pane down value set to 1 so that it only shows the first item of each category. I then unchecked "show header" for "item" so that the item detail doesn't show
