SSWUG.Org Virtual Workshop: Query Plans and Processing
This Friday from AM – AM PST
Here’s your opportunity to get a handle on making your queries work for you. Kalen Delaney, a name well known for things SQL Server will be hosting this workshop. That alone makes it worth the time.
This is definitely the real thing. Kalen provides authoritative, in depth understanding about what makes your queries work, and how to improve performance with the tools available right out of the box.
As a bonus, you can complete a post course quiz and you receive a certificate of completion. While you’re looking at how to best optimize your training expenses this year, consider registering for more than one virtual conference this year, and receive a 20% discount.
Hope to see you there.
Select Views – Stephen Wynkoop, Friday, January 7, 2011
More, smaller databases? Manage your checklists, our featured blog of the week for business intelligence…Also, the hotmail outage and SQL Server, 3rd party tools and more.
[Watch the Show]
Extend Your Membership
Extend your membership or help a friend sign up for SSWUG at a discount with the VIP code "SSWUNL1111". This offer is only available to the first 100 users and only during the month of January.
Featured Article(s)
Tips for using SQL Server 2008 tempdb database
In this article, you can find some useful tips to use SQL Server 2008 tempdb database.
Distributed Data – Opportunity for MS?
Microsoft has been influential in the software industry in finding ways to make significant impact to business capabilities.
When the big boys were out there saying that PCs would never amount to anything in the business world, Microsoft was one of the companies that made it happen.
I did database work for two large companies, both having IBM and HP mainframe bread and butter applications. But when they decided to get their feet wet with new technology, they turned to Microsoft SQL Server. Either company could have used Sybase, Oracle, or even DB2. But SQL Server was the perfect fit for new technology.
SQL Server was the Quicken of relational databases. It wasn’t free; it wasn’t cheep; but, it was the best performance per dollar available.
We haven’t been totally left behind. There is still SQL Server Express and SQL Standard. But with the recent release of Denali, it is clear where the Microsoft focus appears to be. They want the BI world…and we really need that.
In the mean time, the relational database engine scene is paying a price. No enhancements worth mentioning for 3 years. More features have been moved to the Enterprise edition, and the cost has increased for what is left.
Here is my thought…make SQL Standard work like the parallel data warehouse in a distributed form. Most of the current distributed data storage today is not relational. Is that a requirement for distribution? I don’t think so, if it can be done with Parallel Data Warehouse.
Here are a couple responses from last week that have prompted my rather probing question…that and the fact that two key executives on the SQL Server/BI teams at MS are leaving.
Jonathan shares his experience with a Roll-Your-Own distributed database.
I have in my personal digital asset store about 5TB of data; many small files but most of the volume in massive image and audio files. I also have a fast-growing relational set of databases. I’m in the process of creating a distributed SQL Server Express store using Filestream and many different databases to store relational and file data together, distributed across multiple laptops and desktops at various sites; dirt cheap and slow performance.
My goal is archival storage of the data as well as sensational search and retrieval using all the SQL Server tools.
Another feature is the local db has the data set needed for the local user; but access to the larger data set is by getting it from the remote or nearby computers and their arrays of databases and portable digital storage devices. (Of course SQL Server Express limits you to 10GB of non-file-stream data per db.)
Updates are by network connections as well as by taking portable storage devices from computer to computer; either on is valid allowing totally disconnected network segments totally off-line as well as reasonable-speed communication.
I don’t know if I will succeed, but I think that it is possible and I’ve got a start on it
Michal represents a company with a product called SQL Parallel Boost. I am including his response not as an endorsement for his product, but because it is the kind of thing that could take the SQL Server product to the next level of parallel performance.
With the increasing Hardware-power of servers, parallelization is the magic word to unleash the full performance potential.
As DWH/ETL environments on SQL Server platforms, unfortunately don’t parallelize high volume data modification operations (like Oracle)
like regular query operations (SELECT).
We’ve developed a pure T-SQL solution to parallelize UPDATE/INSERT/DELETE on SQL Server platforms thru an own
splitting/dispatching logic with fully automated execution of parallel running tasks and returning the consolidated execution status.
The resulting performance gain is about 5-10 times compared to regular execution.
There is a 2 page brochure available for download on www.ibax.ch.
So there is a thought, Microsoft. You guys are capitalizing on scale out and parallel processing. So, why not increase your performance by scaling out with a truly parallel database system. Why not scale with more smaller machines rather than a Data Center. Doesn’t this actually work better in a cloud?
Well, I hope that gets you thinking at the start of this week. Feel free to sharpen this brainstorm. Send your comments or corrections to btaylor@sswug.org.
Cheers,
Ben