Opinion: Case Sensitivity is a Pox on Computing

Case sensitivity in comparisons is an aspect of computing that I’m surprised is still so widespread. I can’t say it more clearly than this:

It’s a pox on computing and needs to be eradicated.

(https://www.phrases.org.uk/bulletin_board/50/messages/583.html)

I’ve recently been working at a site at present where a new case-sensitive SQL Server system is being implemented. I cannot begin to describe what a poor idea I think this is.

In the end, all that a case sensitive system allows you to do is to have:

  • Multiple identifiers exist
  • They exist in the same scope
  • The names of the identifiers differ only by case

You’d have a hard time convincing me that that would ever be a good idea.

At least not for a system used by typical humans. No sensible person is ever going to be comfortable with “John Smith” being a different name to “john smith”. And similarly, do you really want a single database table with a CustomerID column, a CustomerId column, and a customerID column?

Well I certainly don’t.

In the new system that I mentioned, there are columns ending in ID, Id, etc. They haven’t even been consistent in the naming of their case-sensitive objects.

And yes, I hear the “but what about private variables vs properties in languages like C#?” complaint:

  • Age is the object’s property
  • age is where the object stores the value assigned to the property

Surely we can come up with a better naming convention than that. I’ve lost count of how many times I’ve seen people using the property when they meant the private variable or vice versa. It’s just not sensible.

Now before I hear complaints that case matters, be clear that I’m not talking about case preservation; that’s an entirely different thing. Yes, if I defined a column as CustomerName, I don’t care if I query it by customername, Customername, etc. I want it coming back as CustomerName ie: however I defined it. Case preservation is a virtue; it’s case sensitivity that I see as an almost always painful and unnecessary thing.

Worse, if you’ve ever tested an application against a case-sensitive server, you’ll understand the challenges involved. It’s hard to get case-sensitive code correct.

I gather that SQL Server Management Studio has a current bug that arises when you remove and re-add a database from an availability group on a case-sensitive server. Why? It appears that they aliased a table with A in one place, and used the alias as a in another place. It’s really nonsense to have a situation where that matters but it highlights the other big issue. It makes for fragile applications.

Image by Michał Parzuchowski

Do you really want to be the one who’s testing all your applications and 3rd party utilities to find out if they’ve tested case-sensitivity properly? Do the tools that you use have a better testing regime than SSMS? I’ll bet that most don’t. And what that means is that you get to spend your life wading through obscure tooling issues.

No thanks. Life is too short.

SQL: Do I Still Need to Run DBCC CHECKDB?

In short: YES

(In contradiction to Betteridge's Law of Headlines)

Every now and then, customers ask me if they really need to run DBCC CHECKDB. There was even a question that came up about this on a private mailing list full of people who really should already understand why. There is no mystery here.

DBCC CHECKDB checks the database for physical readability (are the pages intact and can they be read from the I/O system). This makes people wonder that if the I/O subsystem is already doing this, why does SQL Server need to do this?

Importantly, DBCC CHECKDB checks for logical consistency as well.

That’s incredibly important.

SQL Server is a fabulous database but it has no way of coping with internal structures that are messed up. For example, links on pages to other pages, but now pointing to the wrong pages.

Fortunately, because SQL Server is such a good database, having to deal with logical corruption caused by the database engine messing up is a truly rare event. That could happen though if there was a serious bug in SQL Server.

But even if SQL Server is perfect, it could be led astray. A good example would be a multi-path I/O system. Imagine this scenario:

  • The database has two paths to the I/O subsystem (multi-path I/O)
  • It writes a page out one path
  • It reads the same page from the other path
  • The I/O subsystem returns an older version of the page from its cache, not the one that was just written

That sort of thing could easily lead to internal corruption of the database. So could issues with your memory subsystem.

Fortunately, the SQL Server team again have your back. It is designed to help to detect some of these issues too. It probably can’t fix them but it might help to find them. For example, in the multi-path I/O problem mentioned above, SQL Server keeps a list of checksums of recently-written pages. If it reads one of those pages and it gets a page with a valid checksum but a different checksum to the one that it recently wrote, it lets you know that something went very wrong.

Even cleverness like this though isn’t going to help if the culprit is the storage subsystem having a cache bug, and so on, and so on.

Finally, DBCC CHECKDB is so important because it lets you find issues quickly. Otherwise, the logical problem would get much, much worse and be difficult, if not impossible to correct. You don’t want to be the DBA who has a large number of backups but every one of them has the same corrupted data.

SDU Tools: Find columns that shouldn’t have time

Ever since I started working with SQL Server back in 1992, the #1 requested feature that I kept hearing about was a separate date data type. SQL Server had a smalldatetime and a datetime but these included both date and time within the same type.

Having date and time often led to odd bugs or performance issues, where people didn't realize that times were included and tried to work with the values as dates.

In SQL Server 2005, the date data type was meant to be the poster-child for the new SQL CLR integration and there was a shiny new date data type built using it. But when we (the MVPs) looked at it, we all disliked it immensely. The main problem was that it didn't work with all the existing date-related functions. So the date data type was removed from the product before it shipped.

In SQL Server 2008, we got new intrinsic (standard built-in) data types for date and time (and a few interesting other ones), and they worked with the date-related functions. This was what was needed.

However, there's a lot of code out there that was built before this data type existed, or was simply built with the wrong data type.

As part of our free SDU Tools collection, we have provided two ways to find columns that might be worth changing:

ListPotentialDateColumns looks at the names of the columns. If the names suggest that it's a date, but the data type is one that includes time, it'll flag the column as a potential. You can see it being used here:

In each case, the column name suggests that it's a date, but the data type says otherwise. That runs quite quickly as it's just looking at the column names and data types.

Harder to find are columns that are declared with a datetime, smalldatetime, datetime2, etc. yet don't contain any time values. To make these earlier to locate, we've built another tool.

ListPotentialDateColumnsByValue looks through all the values in a column. If the column only contains dates, it reports it:

Now that's obviously a bit slower as it has too look through all the data. After running this (for example), we know that the ValidFrom column in Sales.BuyingGroups only contains dates, without any time component. It's likely that it should have been a date column instead.

You can see both of these in action here:

You'll find more information on our free SDU Tools here:

http://sqldownunder.com/sdu-tools

 

Opinion: Avoid Unneces Abbrevs

Many database developers (and other developers) seem to regard the endless use of abbreviations as some badge of honor. Don't be one of these people.

Avoid abbreviations almost all the time.

I've written before about my dislike for the EOMONTH T-SQL statement. Given the same version introduced names like DATETIMEOFFSETFROMPARTS, surely we didn't have to save 3 characters and could have had ENDOFMONTH. (I heard it was named this way to match the Excel function but matching something from another language that was created a long time ago isn't the right answer here).

I'd also note that I'd rather have seen it as END_OF_MONTH. It's a real pity that T-SQL has names like DATETIMEFROMPARTS but also names like NODE_FROM_PARTS all introduced in recent years. Again, I think there's a need for more cranky old dudes in the development team.

Now we can't change the language directly, but we can make better decisions when using the language.

For example, when using the DATEADD function, we have two choices for parameters. One is a set of (sometimes fairly obscure) abbreviated values. The other is a set of actual English words.

Please choose to use words not abbreviations.

Don't write this code like this:

DATEADD(d, 1, @StartDate)

DATEADD(dd, 1, @StartDate)

DATEADD(y, 1, @StartDate)  

DATEADD(ww, 12, @StartDate)

Are you confident about what each of those means? Instead write this:

DATEADD(day, 1, @StartDate)

DATEADD(day, 1, @StartDate)

DATEADD(dayofyear, 1, @StartDate)

DATEADD(week, 12, @StartDate)

The same applies to column names. Don't write Mbr when you could have written Member.

 

SQL: Newbie Mistake #1: Using float instead of decimal

When I’m looking at a database schema for the first time, there are a number of tell-tale signs that give me the hint that the developers really haven’t done much work with SQL Server before. They’ve made a “newbie” mistake.

One of those is the extensive use of the float data type.

Most times that I see this, the developers have come from a C or Java background and they assume that something that needs a decimal point in it, needs to be float. There are some situations where float makes sense, but 99% of the time what they should have used was decimal.

float is used to store approximate values, not exact values. It has a precision from 1 to 53 digits.

real is similar but is an IEEE standard floating point value, equivalent to float(24).

Neither should be used for storing monetary values. Because the values cannot be stored precisely, people who use float end up with values that don’t match, columns of values that don’t quite add up, or totals that are a cent out, etc. They spend their lives trying to round values to fix the issue, and usually don’t get it right.

Image by Olga DeLawrence

Here’s an example. What values should this code print?

You’d expect the values 0.0, 0.1, 0.2 and so on up to 10.0. But that’s not what you get. The query would continue until the maximum value for the data type was exceeded (a long time). If you stop the query, you’ll see odd values:

Worse, note that our stop value of 10 is actually shown, but it didn’t stop:

The problem is that while the value 10 can be stored accurately in float, the value 0.1 can’t be. In decimal, we have recurring fractions. 1/3 is 0.33333 recurring. We can’t write it precisely in decimal. In binary though, 0.1 has the same issue.

So even though we had a test of WHILE @Value <> 10.0, the value never exactly equalled 10.0. So why does it show 10 in the Messages tab? That’s because SQL Server Management Studio (SSMS) rounds the values that it prints. It’s not showing us the actual value.

We could fix this by substracting @Value from 10 and taking the absolute value of the result, then comparing it to a small increment. But who wants to write code like that?

In SQL Server, decimal, numeric, money, and smallmoney are the data types with decimal places that store values precisely. numeric is basically a synonym for decimal. money and smallmoney are old Sybase data types that have fixed scale, and have a funky relationship with currency symbols when converting strings. I generally don’t use those. (There are some arguments for them in gigantic data warehouses where their smaller storage size might help but with row compression, the reasons for that are quickly disappearing). It’s one of the problems with backwards compatibility in SQL Server.

You can’t blame people for using a data type called money for storing amounts of money. But it’s generally not the right answer.

Let’s now look at the query from before if we change to decimal:

When executed, it stops exactly as expected:

Decimal (and numeric) require a precision and a scale. These should be chosen appropriately to store the values that you need. You need to keep rounding in mind when you calculate decimal values.

As I mentioned earlier, there are places where float and/or real make sense, but they are typically scientific calculations, not business calculations.

 

 

 

SDU Tools: Listing all the Data Types Used in a SQL Server Database

 

When I first start to work with a database that I haven't seen before, one of the things that I check immediately is the type of data being stored. The list of existing data types often gives me interesting insights into the design of the database, and how it interacts with application code.

The list of data types also shows me details of the consistency (or lack of) in the design.

In the image above, you can see the data types that I used when creating the WideWorldImporters database samples for Microsoft. This listing was created by our ListAllDataTypesInUse stored procedure in our free SDU Tools suite.

For another example of it's use, let's take a look at the AdventureWorks sample database, and see what it shows us:

The parameters to the procedure are the name of the database to check, and a list of schemas, tables, and/or columns to limit the search to. For each one, you can provide the word ALL, or a comma-delimited list of values. For example, @SchemasToList = 'HumanResources,Sales' would limit the output to tables in the HumanResources and Sales schemas.

In this case, you can already see a few things. First is that user-defined data types have been used (note AccountNumber in row 1, Flag in rows 9, Name in row 14, and NameStyle in row 15).

Decimal values have a precision of both 8 and 9, and scales of 2 and 4. That's often interesting to see how decimal values are stored, and how rounding is being used in the application.

I'm not a fan of the money data type (it's an old Sybase type), but note that it is used, as is numeric, along with the decimal values. That would have me  a bit curious.

Obviously, I'd be more concerned if I found text, ntext, or image data types, as they are deprecated.

I tend to also rarely use nchar, so I'd be a bit interested in what those values are for.

This part of the list let's me get an idea how string values are stored. It also lets me start to find issues like email addresses or product descriptions that are stored in different size columns in different parts of the database.

And finally I get to notice some more user-defined types, plus smallmoney (likely a red flag for me), some GUID values, and some xml.

You can see it in action here:

For more information on our free SDU Tools, please visit:

http://sqldownunder.com/sdu-tools

 

 

 

 

SQL: Best Way to Scale SQL Server Database Performance

I see so much written about how to scale SQL Server systems, and this generally starts with needing to improve SQL Server database performance. When I read articles from the SQL Server field support teams with titles like Top 10 Performance Problems for SQL Server, I often just smile.

The problem is one of perspective. If you are looking at the performance problems that are brought to the support teams to solve, you get a very, very skewed view of what's typical.

That's because most performance problems are dealt with by database developers, DBAs, application developers, etc. long before they'd go anywhere near product support. So the product support team never sees the most common problems.

If I had to rate performance problems by cause, I'd suggest the following:

  • 40% of issues are related to poor application design
  • 25% of issues are related to poor indexing
  • 20% of issues are related to how the database framework that the application has used talks to the database (either by design or by how it's been configured)
  • 10% of issues are related to poor concurrency design (blocking issues)
  • The remaining 5% are odd things, and generally the ones that go to product support.

But if I had to provide my #1 tip for improving database performance and increasing scale, it's this:

Stop talking to the database!

It's that simple. So many applications that I work with are very, very chatty.

A year back, I stepped through the login process of a web application. A single login ended up causing over 400 calls to the database. That should probably be one or two.

I've seen processes that send 55 million calls to the database and run in half an hour. But the same process runs in 2 minutes when the number of database calls is cut to 4000.

I've seen windows apps that execute 90,000 remote procedure calls before the first screen comes up. (Loading 90,000 customers one row by agonizing row at a time). It's a tribute to SQL Server that the application started in about 30 seconds, but that should have been one call, not 90,000.

I've recently seen frameworks that have been misconfigured. For every select of a single row of data, the framework executed this:

  • Create a cursor for the command
  • Open the cursor
  • Fetch from the cursor
  • Fetch again from the cursor (this one failed)
  • Close the cursor
  • Re-open the cursor
  • Query for the parameter metadata
  • Close the cursor

And that was done every single time that a single SELECT should have been sent. The problem is that the developers using frameworks like this are blissfully aware of the commands being generated for them in the background.

I've seen caching errors where an application ran the same query with the same parameters over 12,000 times per minute. The developers were unaware of this. (They had a typo in the caching code but it "looked like it worked").

The bottom line is that you need to spend time looking at how your applications interact with the database. If you execute 400 commands every time someone logs on, you will never get any real scale.

I'll say it again. The best way to scale a SQL Server database (and to get better performance from it) is to stop talking to it incessantly.

 

SDU Tools: Determining Leap Years in T-SQL with IsLeapYear

Is 2018 be a leap year?

Was 2000 a leap year?

Will 2100 be a leap year?

When I was a young student at school, we learned that leap years were every four years. Then as I got older, I learned that there was more to it than that.

It's important to understand that any calendar is just an approximation, and there is a lot more to calendars than most people realize.

For those that would really like to get some background, my favorite calendar-related site is Claus Tøndering's Calendar FAQ. It's a wonderful collection of details of calendars and their history. In addition, for those that are into coding, he has sample program code to determine things like when is Easter next year?

Currently the rule for determining leap years is:

  • Every fourth year
  • Except if the year is divisible by 100 but not by 400

So, to answer the questions above, 2018 is not a leap year, 2000 was a leap year, but 2100 is not a leap year. (It's the last one that usually surprises people).

Apparently the scientists are still discussing where or not there should be a 4000 year rule. Can't say that I care as I won't be here.

We added a simple tool to do these calculations:

You can see it in action here:

For more information on our free SDU Tools, please visit:

http://sqldownunder.com/sdu-tools

 

 

 

 

Shortcut: Using Colors to Avoid Running Scripts Against the Wrong Server

Everyone who's worked with SQL Server for any length of time, has had the experience of executing a T-SQL script, and then noticing, with horror, that they've just executed the script against the wrong server.

You know the feeling. It even happens at Christmas time, just when you were hoping to get away from work for a few days, or when you are the unlucky one who's doing on call work.

Many of these surprises would be avoided if there was something that gave you a visual clue that you were connected to the wrong server.

SQL Server Management Studio (SSMS) has had an option to color code connections to servers for quite a while. The solution isn't perfect and isn't as good as Mladen Prajdić's SSMS Tools Pack:

http://www.ssmstoolspack.com/

(If you're keen to pay for an add-in, I do recommend that one)

For many people though, the colorizing provided by SSMS is just enough. And it's easy to use.

When you open a new database connection, you can click Options:

This opens a larger dialog that allows you to specify a color for the connection:

I've chosen Red to warn me that this is a production server.

Then, you'll see the color bar at the bottom of each script window in SSMS:

 

 

 

 

SDU Tools: Show SQL Server Backup Completion Estimates

When you first start working with SQL Server, you're often working with smaller databases and every backup seems to happen quickly. SQL Server is fast at creating backups.

When you get to larger databases, backups start to take a while. Fortunately, backup is one of the commands that sets a percentage complete value that can be seen in the sys.dm_exec_requests system view.

And when the databases get even larger, the question quickly becomes:

When will my backup finish?

The good news is that we've built a tool to make that easy to work out.

One of the SDU Tools is ShowBackupCompletionEstimates. It shows you any currently executing backups, how complete they are, when they started, how long they've taken so far (in seconds), and when we estimate that it will complete. Because it can cope with more than one backup running at a time, we also added the last SQL statement that was executed in the session, to make it easier to work out which is which.

Here's an example of its output (wrapped for clarity):

You can see it action here:

For more information on our free SDU Tools, please visit this site:

http://sqldownunder.com/sdu-tools