Databases Reference
In-Depth Information
I can think of one. For example, we have an initial QlikView document with the
Airline Operations
data (the one we constructed in
Chapter 2
,
Seeing is Believing
).
Based on this initial data model, we can create a new QlikView document, to
which we will only add the
Employment
data (used in
Chapter 8
,
Data Modeling Best
Practices
). As a result, we would have one
Airline Operations
document accessed
by certain users, and another with the same data but with additional information
about airline employment, which might be treated as confidential and accessed by
another group of users.
As a side note, we must point out that binary loads are used in yet another approach
to data architecture. We will not discuss it in-depth, but suffice to say that the new
layer is composed of QlikView documents consisting of only a data model without
any frontend objects, referred to as "QlikMarts". These QlikMarts then become the
source for the QlikView documents in the Presentation Layer.
Loading from RAM
In some cases, we will need to read the same table more than once in a single script
execution. This means, querying the database (or QVDs) and pulling data from it,
and then reprocessing that same data after the first read in order to make it adequate
for our data model. Since the data is being stored in RAM after each query during
the script execution, we can use that RAM-stored data instead of going directly to the
original data source. This is accomplished via a
Resident load
.
Resident load
The keyword
Resident
can be likened to the keyword
From
in a query. The
difference is that the
Resident
keyword is used to reference the data in RAM model,
that is, all the tables that have been previously read in the preceding queries of the
same script. The process for achieving this is as follows:
1. First, we must load data from a data source (any database or table file
described in the previous sections), so we create the corresponding query in
the script. An example would be:
SalesData:
LOAD
InvoiceNumber,
Date,
SalesPerson,
Department,
Amount as InvoiceAmount;
SQL SELECT * FROM DataBaseName.dbo.Sales;
