Data Warehouse with Multiple SQL Requesters
Last, let us look at an example that is probably realistic for most businesses. We have one datastore for
the enterprise that consists of several data servers and a mainframe, collectively called a Data
Warehouse or a Distributed Data Processing system. That conglomerate of a datastore receives SQL
requests for internal business operations from a business logic server. The business logic server is
processing operations as directed by employees using PCs and front-end software on their desks.
The company also has a website serviced by a web server. The web server interacts with visitors and
then sends SQL requests to the same enterprise data warehouse. Although there are many brands of
software involved and many platforms they are all able to speak and understand SQL.
We can trace the information flow in this graphic by starting with the PCs on the right. They are
connected to either the Department Server or the Web Server. These servers can then create a SQL
request using their business logic software and pass the request to one or more of the Data Servers. This
is usually done through a high-speed LAN which only connects servers (represented on the far left).
After the SQL Request has been processed the results can return from left to right through the servers
and to the PCs.
(Click image for full size)
