Looking forward to "Yukon"
roundup press release confirms that the next version of SQL Server,
code-named "Yukon" is still on track to ship in the second half of 2004. With a
lot of development shops slowing down for the end of the year, this is a good
time to spend a few minutes looking forward to "Yukon".
Don't get me wrong. For most developers, it's not a good time to be writing
code yet. The early beta of SQL Server "Yukon" that was handed out at the PDC
will undoubtedly undergo some major changes before the product is released. You
certainly don't want to be doing detailed design of production systems based on
this beta. But what you can do is get a sense of the major innovations in
the software that might affect your own workflow. As we move through 2004 and
plan new applications, you can start basing your plans on these new features,
and when the software is released, you can be ready to focus your efforts on the
pieces that are important to you.
Of course, not every developer and DBA will focus on the same parts of this
big, sprawling product. But to get you started, here's my own
perhaps-idiosyncratic list of key "Yukon" features to start considering as you
look forward. I'm concentrating on developer-oriented features; DBAs get the
expected "bigger, better, faster, stronger" improvements too, but that's a
topic for another article.
Managed code calls from T-SQL. You'll be able to use any of the .NET
languages to write functions, stored procedures, and triggers to be called
directly from T-SQL. Ever wanted to do something in T-SQL, but discovered that
the language wasn't rich enough, the performance wasn't good enough, or access
to machine resources wasn't wide enough? This could be your answer.
The xml data type. This is more than just stuffing XML into an
nvarchar column. In particular, when you define a column to be of type xml, and
store XML documents in it, you can then use XQuery to extend the T-SQL query
syntax to searching within the XML; it's not just a featureless blob. Microsoft
is also extended XQuery to handle other operations within the XML, such as
inserts and updates.
Reporting Services. SQL Server is finally getting a comprehensive
reporting layer that can produce XML, HTML, PDF, Excel, and other outputs. The
report designer integrates with Visual Studio .NET, and the whole solution makes
heavy use of .NET technologies. You actually don't have to wait for "Yukon" for
this piece; the SQL
Server 2000 version is due to ship in the first quarter of 2004.
Service Broker. Microsoft is implementing an entire asynchronous
messaging layer directly in SQL Server, with T-SQL extensions to handle
messaging. Messages are delivered reliably and in order, even if systems are
only occasionally connected or get rebooted during the process. If you've been
lashing together other technologies to make distributed database systems work
across the Internet, this might be a great boon to you. Service Broker also
enables query notifications, which let you subscribe to a query and be
notified when its results change.
ADOMD.NET. If you're a user of SQL Server Analysis Services and do
your work in .NET, you'll be pleased to know that there are now native managed
objects for interacting with Analysis Services.
SQL Management Objects. SMO is the successor to DMO, offering even
more power for programmatically managing SQL Server. New features include
capture and batched execution of T-SQL, WMI management for SQL services, and
.NET integration. There's also a separate set of Repliation Management Objects
(RMO) for handling replication from code.
SQL Server Workbench. This is going to be one of those features that
some people love and some people hate. Gone are the independent tools like Query
Analyzer and Enterprise Manager; instead, all of the client functionality for
managing and interacting with SQL Server is crammed into a version of the Visual
Studio .NET shell named SQL Workbench. On the plus side, experienced VS .NET
developers will have their productivity tools available for SQL Server tasks. On
the minus side, experienced DBAs will have to learn new ways to do things, and
some features (like database diagrams) are simply vanishing because they
couldn't be ported easily.
As you can see, there's a lot coming down the pike - and I haven't even
mentioned persisted computed columns, full-text indexing enhancements, the new
DTS architecture, improvements in data mining...and a host of other things. As
you see more and more "Yukon" articles over the coming months, try to keep the
big picture in mind and focus on those parts of the new software that you can
use to directly improve your own applications. That's the best way that I know
of to get started with a huge new technology.
Mike Gunderloy has been developing software for a quarter-century now, and writing about it for nearly as long. He walked away from a .NET development career in 2006 and has been a happy Rails user ever since. Mike blogs at A Fresh Cup.