You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@hbase.apache.org by Software Dev <st...@gmail.com> on 2014/05/03 00:57:09 UTC

Designing an increment table

We have a table that will house a bunch of increments for our
reporting needs. We will be saving things as number of page views,
searches, interactions, etc. We would like to store attributes along
with these increments such as sex, status of user, country of origin
etc. Would we would like to do is graph this out and then filter out
by any of these attributes. For example

- Over the last 3 weeks, how many page views were from the United States.
- Over the last 3 weeks, how many page views were from the United
States where the user is a male.
- Over the last 3 weeks, how many page views were from the United
States where the user is a male and he was logged in
- Over the last 3 weeks, how many page views were from the United
States where the user is a male and ....

How would I go about designing this table so these questions could be
answered? We will always be including a date range as well as country.
The other attributes such as sex, logged in status may or may not be
included.

Is there a Fuzzy type of filter for columns? If so I think I could
just suffix all the columns with a byte array. The first byte can
denote the sex of the user, the second the status, etc. Would that
work?

Any suggestions?