Database Reference
In-Depth Information
Why Is PDW Important?
PDW is actually branded
SQL Server Parallel Data Warehouse
. This leads
to misconceptions about what PDW is or isn't. Actually, it'd be more
accurate to say it is Parallel Data Warehouse, powered by SQL Server. There
is an important but subtle difference. SQL Server is a mature utilitarian
database engine with some enhanced functionality aimed at improving
performance for the data warehouse workload. It has a myriad of features
and options to support other workloads and products as well. PDW is
different. PDW is designed purely for data warehousing. It is built as a
black box solution that is designed to streamline development, improve
performance, and reduce the burden on operational teams.
However, here is a different way of looking at PDW:
“
PDW is the de facto storage engine for data warehousing on the
Microsoft Data Platform
.” —
James Rowland-Jones
,
circa 2013
I realize that I am quoting myself. It's not my intention to be egotistical.
However, I do believe it to be true, and I don't know of anyone else who
has come out and publicly said anything to the same degree. Besides which,
I want to share with you my reasoning. I don't know when I started using
it exactly, but it is something that has resonated well with attendees of the
PDW training I've given, and I hope that it will resonate with you too.
To help back up my somewhat grandiose hypothesis, I want to draw your
attention to the following key points, which I explain more fully in the
following sections:
• PDW is the only scale-out relational database technology in the
Microsoft Data Platform.
• Data warehouse features come first to PDW.
• PDW is the only Microsoft relational database technology that
seamlessly integrates with Hadoop.
• PDW has two functional releases a year.
PDW Is the Only Scale-out Relational Database in the Microsoft
Platform
Some attempts have been made at scale-out with the likes of federated
databasesanddistributedpartitionedviews,butitisalsofairtosaythatSQL
