Welcome to this month’s T-SQL Tuesday #118 roundup. Now, I tried to publish this as soon as possible. Believe me it was no easy task after all the great contributions this month.
Personally, I’d like to say a big thank you to everybody who contributed this month. In addition, Steve Jones who allowed me to host this month.
Especially since I know you all have other commitments and there were other SQL things happening during the last week.
I’d also like to mention that there were at least four first-time T-SQL Tuesday contributors this month. Including Jeff Miller, SQLDork, Eitan Blumin and Rob Litjens.
I hope they had as much fun contributing their posts as I had reading them afterwards. For those who missed the invitation you can read about it here.
For some reason I thought it’d be a good idea to do it on my birthday month when I first given choices about which month to host. However, now I cannot remember why.
Now, some of you didn’t follow all the rules this month. However, I have decided to let that slide so just remember to follow them properly next month.
Therefore, below is a roundup of all the submitted posts. Because they were 23 contributions this month, I have also provided links below to each one. Furthermore, they are in the order that I saw them published.
- Greg Dodd
- Reitse Eskens
- Rob Farley
- Alexander Arvidsson
- Kevin Chant
- David Fowler
- Bert Wagner
- Jon Shaulis
- Brent Ozar
- Jeff Miller
- Kevin Hill
- Kenneth Fisher
- SQLDork
- Andy Levy
- Erik Darling
- Jess Pomfret
- Eitan Blumin
- Rob Litjens
- @sqL_handLe
- Shane O’Neill
- Glenn Berry
- Nate Johnson
- Matthew McGiffen
Posts
Greg Dodd
Greg talks about how they’d like to see a foreign key that references more than one table. Within the post is also a good example about what is being explained.
It’s certainly something to think about. Especially when it makes you think about which alternatives you must do instead.
You can read that post in detail here.
Reitse Eskens
Reitse discusses how they would like an estimate on how long SQL Server would take to process data. Which could be done by either time in minutes or by cups of coffee instead.
My burning question about this is coffee from where though? Because coffee from a vending machine is far quicker than from a superior machine.
You can read that post in detail here.
Rob Farley
Rob talks about something that they have been wanting for many years. In fact, it’s something they raised a Connect item about previously.
Which is about making more predicates SARGable. Because I am keen on performance tuning, I really like this idea.
I’d just like to point out that this is Rob’s 118th T-SQL Tuesday contribution. In fact, in two months time they’ve contributed for T-SQL Tuesday every time since the beginning ten years ago.
Anyway, you can read their post in detail here.
Alexander Arvidsson
Alexander shares their idea about being able to use a WHERE statement with TRUNCATE. I must admit that I really like this idea.
In reality, Microsoft did a similar thing with indexes which is how filtered indexes came to be. So, I think this would be a very welcomed addition to SQL Server.
You can read this post in detail here.
Kevin Chant
I discuss my idea about the automatic index management feature having multiple levels. Which would allow you have a more aggressive indexing strategy.
For example, it could create more complicated indexes and introducing Columnstore Indexes more often. In addition, I do mention the potential risks of having this.
You can read my post in detail here.
David Fowler
David talks about the possibility of having object level restores, which includes individual tables as well. They admit it’s possible with third party tools, however it would be good to see this in SQL Server.
In addition, it’s worth noting that after David’s contribution was published Brent Ozar contributed with the same suggestion. So, there is obviously a desire for this.
You can read that post in detail here.
Bert Wagner
Bert definitely deserves a special mention here. Because as well as doing a T-SQL Tuesday contribution they’ve uploaded a video to YouTube about it which you can watch here.
Anyway, the post and video are about having a native feature which audits who last modified a row and when. Having used a couple of alternatives that are discussed here myself I can see why and agree it could be of use.
I recommend watching the video, however you can read the post instead here.
Jon Shaulis
Jon discusses three features that they would like to see in their post. Entity Relationship Diagrams, Database Documentation and Culture Change driven by technology.
For me, their thoughts about whether SQL Server could introduce culture change in the future certainly provoked a lot of thoughts.
Anyway, you can read the post in detail here.
Brent Ozar
I’m certain that Brent now has a new T-SQL Tuesday record. Because Brent announced their post was completed within couple of hours of the invitation going out.
Being a person of high profile, this was bound to be highly anticipated, and Brent delivered with his simple idea of being able to restore objects.
If it sounds familiar it’s probably because David Fowler coincidentally suggested the same thing earlier in the day. Obviously, they’ve both experienced issues because of this.
You can read Brent’s post in detail here. Afterwards, I recommend clicking on the link to Gonzalo Bissio’s original feature request if you want to see it happen.
Jeff Miller
Jeff is a first-time T-SQL Tuesday contributor. Hopefully the first of many more in the future.
In this post Jeff has an interesting suggestion relating to column formatting. Which was inspired by PowerQuery used in Power BI.
You can read the post in detail here.
Kevin Hill
So, this month two of the T-SQL Tuesday contributors are called Kevin and are cyclists. In addition, this is the first of three T-SQL Tuesday contributions from the Dallas DBA team.
Rather coincidentally, Kevin asked a question similar to this topic on LinkedIn a couple of weeks ago.
Somebody asked me about load balanced writes many years ago. In fact, somebody named it split stream and the term has being stuck in my head ever since.
Like Kevin says I think the other idea is covered by the DBATools PowerShell module. You can read the post in detail here.
Kenneth Fisher
Kenneth has a couple of ideas in their post. First is about sp_helpindex and the second is about better SQL Agent job permissions.
I really like the second one about better SQL Agent job permissions. Because I think there is some merit in it.
You can read the post in detail here.
SQLDork
In truth, I don’t know what their real name is. However, I do know is that they work with Kevin Hill. In addition, I also know that it’s their first T-SQL Tuesday contribution.
Within their post they share their pain about diagnosing SSIS issues.
You can read that post in detail here.
Andy Levy
Andy talks about how they want a way for SQL Server to query multiple databases at once and merge the results together. In addition, they want these queries to run in parallel.
It is mostly aimed at the on-premises version of SQL Server. Personally, I can see some business cases for this feature. Especially since there are so many on-premises installations of SQL Server that still exist and probably will for some time.
You can read this post in detail here.
Erik Darling
Erik suggests their idea about being able to edit execution plans. After editing your plan, you could then force SQL Server to use it.
Having dealt with issues related to execution plans for many years this could be an interesting option. Imagine, instead of having to tune SQL Server to try and get the perfect plan you simply change it.
Of course, like Erik highlights you could end up making things worse. Still, something to consider. You can read that post in detail here.
Jess Pomfret
Jess discusses two ideas in their post, and as far as I am concerned, they are improvising instead of cheating for suggesting them both.
First suggestion is merging the LIKE and the IN operators to have a combined LIKEIN operator. I like the concept because theoretically it could improve queries.
Second suggestion is the ability to set SQL Server Agent to not send an alert if the message contains certain text. Like Jess says this would reduce noise in certain situations.
Anyway, you can read Jess’s post in detail here.
Eitan Blumin
Eitan is another first-time contributor to T-SQL Tuesday. Hopefully the first of many contributions because I enjoyed reading their suggestions for improvements to SQL Server Table Partitioning.
Within this post are some fairly good ideas to improve it and I think some of you out there will agree. In addition, I hope these suggestions have been submitted to Microsoft.
You can read their post in detail here.
Rob Litjens
Rob is also a first-time contributor to T-SQL Tuesday. I hope this is the first of many for them as well.
Within their post Rob discusses their ideas of improving things like SPN generation and removals.
In addition, Rob also discusses GMSA’s. You can read their post in detail here.
@sqL_handLe
I will use the name given by the contributor for this post. Anyway, this post is brief and straight to the point.
In short, their idea is for meta-data only changes for Clustered Columnstore Index (CCI) partitions. As you can read in the post doing this has various advantages.
You can read that post in detail here.
Shane O’Neill
Shane starts off their post by apologizing because they think their post sounds like they are complaining. After reading the post I can assure there is no need.
Basically, Shane wants SQL Server to have a performance rating. Something along the lines of x out of 100. Considering the next contributor below wants a similar thing for server hardware there may be some merit in this.
You can read the post in detail below.
Glenn Berry
Glenn’s idea for this month is something that they have informally suggested to some people at Microsoft in the past.
To summarize, Glenn suggests that SQL Server has an internal benchmarking utility that measures the performance and storage that it was running on.
Afterwards that data could be used by the Query Optimizer to create optimal query plans. A lot of you know that Glenn is considered a hardware expert in the community, so this idea is well explained.
You can read the post in detail here.
Nate Johnson
Nate talks about how nice it would be if there was a more polished way to do load testing. Within the post they share how hard it is to do this now as well as an ideal way to do it.
In addition, they mention three other people who have contributed to T-SQL Tuesday this month. As an apology for what Nate feels turns into a bit of a rant there’s also a cute picture of a dog at the bottom of the post.
You can read Nate’s post in detail here.
Matthew McGiffen
By their own admission Matthew’s contribution this month was closer to the deadline than others. However, I’m glad somebody living in the South West of the UK contributed because that is where I am originally from.
Matthew has a few things on their wish list. For me, the one that stands out the most is making the Enterprise Edition the Standard Edition.
Some Enterprise only features have made their way to Standard now so one day this may well become a reality.
You can read this post in detail here.
Final word
I hope everybody enjoyed reading and/or writing the contributions this month because I know I did. Hosting has been a fun experience for me and hopefully Steve Jones will allow me to do it again in the future.
In the meantime, the ten-year anniversary of T-SQL Tuesday is approaching so keep your eyes open.
[…] and roundup from Kevin […]