Database Reference
In-Depth Information
How Do These Models Fit into your organization?
Think through your organization and the types of people and roles who are
always working with data, asking for more data, and needing more memory
on their machines or access to data sources. Some examples include:
uu
Accounting and finance teams
uu
Marketing analysts
uu
Retail, manufacturing, and logistics analysts
uu
Executives
uu
Reporting analysts
Many of these people will already be familiar with these tools but will love
the additional assistance professionals can give them with all the new things
you're learning in this text.
CoLuMN STorES
The term
column store
has been mentioned
several times in this chapter already with only
a basic explanation. This section highlights the
importance and building blocks of this powerful
technology. The technology is not new, although
it's very new to client tools such as PowerPivot.
SQL Server's traditional index structure is based
on a B-Tree model. B-Tree models are great for
finding data that match a particular condition in a
query. They are also pretty fast when you need to
scan all the data in a table. There are a couple of
main reasons we need the column store, however.
Compression is critical to getting increased per-
formance out of the same data on the same disks.
Most of the time, data is stored in rows as shown
in Figure 5-5.
With a column store, the data is reorganized into
a column-wise fashion similar to Figure 5-6.
FIguRe 5-5
Row store versus column store
