A Near-Zero Downtime Case Study

I have a post covering an example of making significant changes with near-zero downtime:

This is where we start the decline phase in our story. Our temporary procedures existed as a bridge from the old procedures which took ClientID and new procedures which will take ProfileID. With our final versions of procedures, we replace @ClientID with @ProfileID in the input parameters and update any conditional logic within filters to use ProfileID instead of ClientID.

The only remaining use of ClientID in these procedures is as an output in select statements, as we still need that for the old code; by this time, all of those references are ProfileID AS ClientID. Otherwise, these new procedures are built around ProfileID.

We still have the original procedures that the application code uses and they reference our now-finalized procedures. These procedures are now transition procedures—we need them right now but as we move application code over to the new procedures, we can deprecate these. But I’m getting ahead of myself a little bit.

Click through for the full story as well as a bunch of pictures which completely understate the vastness of work done.

T-SQL Tuesday 112 Roundup

Shane O’Neill has a roundup of T-SQL Tuesday #112:

I’m going to keep this intro short and sweet. I’d like to say it’s because I know these roundups are for acknowledging and thanking the people who have contributed, and this is the case! Mainly though it’s because I wrote this bit last and I’ve already written… checks … 2,795 words!

So thank you one and all for participating. I’ve never thought the posts would raise feelings of happiness, sadness, thoughtfulness, and appreciative-ness that this one has.

Read on for the full list and Shane’s thoughts.

Using Extended Events with Azure Data Studio

Erin Stellato shows how we can use the Extended Events Profiler (oh, how I dislike that name) with Azure Data Studio:

To clarify, the extension in ADS is like XEvent Profiler in Management Studio (which also is built using Extended Events).  The name “SQL Server Profiler” is confusing, as this is not the same tool (UI) that’s been available since SQL Server 7.0.

To install the extension, click on it, and then select Install.  Once it’s installed you can select Reload and it will move into the top half of the window under Enabled.  Notice that when you select the extension, information about how to use it also appears.

Erin has a lot of useful information here, so check it out.

Contrasting Oracle’s Architecture with SQL Server’s

Kellyn Pot’vin-Gorman helps us understand where Oracle’s architecture differs from SQL Server’s:

The first thing you’ll notice is what Oracle refers to as an INSTANCE is different to what SQL Server calls one.

Oracle’s instance is most closely related to what SQL Server calls their database, (although it includes the files that Oracle includes in their description, too) and the Oracle home is *relatively* SQL Server’s version of an instance.

Read on for a deeper comparison from someone who has spent quite a bit of time working with both platforms.

Data Modeling: Not Everything Is VARCHAR(8000)

Bert Wagner gives us several reasons why VARCHAR(8000) can be a bad fit for data types:

When first learning SQL Server, I had the brilliant idea of defining every column on my table as VARCHAR(8000). After all, the VARCHAR datatype only consumes as much storage space as the string itself (plus two bytes), so even if I defined the max length of a column as 8000, I wouldn’t be wasting storage space if my data was smaller than that.

My development life would be easier too since I wouldn’t have to think about the kind of data I was storing in my columns; I could define everything as VARCHAR(8000) once and never have to go back to make any changes. Brilliant!

While I was correct about not wasting storage space, it turns out the idea of making every column VARCHAR(8000) is a terrible idea.

Click through to learn why.

Categories

March 2019
MTWTFSS
« Feb Apr »
 123
45678910
11121314151617
18192021222324
25262728293031