More in Database Performance Tuning

  • May 31, 2012

    SQL Server Database Corruption, Part XII: Recovery Sample 2

    In my last post in this ongoing series on SQL Server database corruption I mentioned that my next post would be to provide a ‘soup to nuts’ sample or example of how you can test corruption and recovery in your own environment – as a means of better getting familiar with exactly what corruption is, what it looks like, and how to address it....More
  • May 16, 2012

    Curious Case of the missing SOS_SCHEDULER_YIELD waits 1

    Many of the monitoring scripts/tools for looking at wait statistics aggregate all the wait statistics and return the top 95% of all waits occurring, sorted by most prevalent waits by total wait time....More
  • Apr 17, 2012

    How to prevent table scans when searching LOB data? 2

    If a SELECT has the LOB column in the WHERE clause then the query plan will involve a table scan – as you cannot create a nonclustered index over a LOB column....More
  • Apr 3, 2012

    What about moving LOB data while keeping the table online?

    Moving LOB data associated with a table isn't simple—even with the SQL Server 2012 ability to rebuild indexes that include LOB data as an online operation. The key reason is the behavior of LOB data during these operations....More
  • illustration of data cube with red arrows indicating movement
    Mar 21, 2012

    What about moving LOB data? 2

    There are actually a couple of ways to move LOB data associated with a table. Let's tackle moving LOB data using OFFLINE operations....More
  • Mar 13, 2012

    Setting permissions on a database mirror database snapshot 2

    The only way to access the data in a mirror database is to create a database snapshot on the mirror database and then have users connect to the database snapshot to run their queries....More
  • Mar 7, 2012

    Do query plans take account of what data is in memory?

    When the query optimizer is evaluating various plans as it narrows down the possible choices for the best plan it can come up with in a reasonable time (which is not always the absolute best plan, as the query optimizer cannot spend an inordinate amount of time doing plan compilation), it makes use of the relational metadata....More
  • table data on white piece of paper
    Feb 27, 2012

    How about Filtered Indexes instead of Partitioning?

    Filtered indexes are an incredibly powerful feature (one of my favorites) so I don't want to dissuade you from using them. A table can be partitioned using the partitioned tables feature. A table can be broken into multiple tables and then unioned (UNION ALL) in a view using the "partitioned views" feature....More
  • computer keyboard Power Shift key
    Feb 23, 2012

    Did SQL Server eliminate any partitions?

    If SQL Server is accessing a table where you don’t think it should, then check these things....More
  • Feb 9, 2012

    Solutions to VLT concerns around statistics and maintenance! 2

    Let's tackle why partitioned views can be a fantastic choice for partitioning large sets—even for new design....More
  • illustration of data with colorful numbers in background
    Feb 2, 2012

    Partitioned Tables v. Partitioned Views–Why are they even still around? 1

    Partitioning is CRITICAL for VLT. What is VLT? It’s about as descriptive as VLDB and it means very large table. Most people speak of VLDBs (very large databases) and they define that as databases that are 100s of gigabytes (many would say that a database that’s 1TB or larger is a VLDB)....More
  • Dec 28, 2011

    Where to store LOB data?

    There are two kinds of data type for storing LOB data – true LOB data types that can store more than 8000 bytes and the limited data types that can store up to 8000 bytes – and of course there are pros and cons for each type....More
  • Dec 28, 2011

    What perfmon counters can I trust when using SAN disks?

    Knowing which counters can be trusted to give you valid information and which counters can't be trusted is very important to properly monitoring and troubleshooting storage and Microsoft SQL Server....More
  • Nov 11, 2011

    How many steps should the histogram have?

    During the final phase of statistics creation (called Histogram consolidation), SQL Server goes through and looks at neighboring steps....More
  • Nov 6, 2011

    Why can't SQL Server update statistics on the fly? 5

    SQL Server stores statistics in 3 parts: the statistics header, the density vector and the histogram. The statistics header has basic (but very helpful) details such as the last time the statistics were updated (column: Updated), the number of rows in the table (column: Rows) vs. the number of rows used to generate the statistics information (column: Rows Sampled) and a few other items such as the number of steps in the histogram (column: Steps)....More
IT/Dev Connections

Register now to get the best rate available!

From the Blogs
computer keyboard with blue save money key
Apr 15, 2014

Save Thousands in Licensing Costs for SQL Server AlwaysOn Availability Groups

Windows Server 2012 or Windows Server 2012 R2 Standard Edition is a great way to save thousands on licensing—and save yourself some serious headaches when deploying SQL Server AlwaysOn Availabiliity Groups....More
Apr 8, 2014

NoSQL: Eventual Consistency Yields Major Flaws

While I believe that NoSQL can have some places where it legitimately makes sense, I also believe that NoSQL is also the patently wrong choice for a number of different types of applications or use cases....More
Apr 1, 2014

SQL Server 2014 Now Available—No Joke

Initial word on the street is that SQL Server 2014 has been very stable so far—which is probably why it went to release so quickly and bypassed what would typically have been another pre-release....More
SQL Server Pro Forums

Get answers to questions, share tips, and engage with the SQL Server community in our Forums.

Sponsored Introduction Continue on to (or wait seconds) ×