SQL Tact

pointers, solutions and scripts for the SQL DBA
Not intended to replace common sense

4/23/2014

Real heavyweights: Float vs Decimal, the Thrilla in Precision

This is a followup to a previous post where I likened SQL Server float datatype to Muhammad Ali. Specifically, I wanted to address the phrase "negligible data storage increase" to the test for different numeric data types.

In short, I will dismiss the notion that float provides a realistic advantage in storage of large numbers. (Also, I had a blog title idea too catchy to pass on.)

I'll avoid recapping in total my earlier post about the dangerous consequences of rounding with the float data type. The float data type is still popular with developers and database designers because of its supposed advantages in data storage size. (Also, some code-first modelers use float as a default. Also dangerous. And lazy.)

Too often, developers and database designers ignore the large potential for inconsistent rounding with float (and real) data types, whereas a decimal(p,s) data type is not only consistent with storage, but it provides transparent logical intent to the development team. No surprises with truncation of numerals to the right of the decimal point with the decimal data type.

Here's my test lab for a number with 3 digits to the right of the decimal point.

create table realtest (real1 float(24))
create table floattest (float53 float)--float(53) is the default
create table decimal9test (decimal9 decimal(9,3))
create table decimal19test (decimal19 decimal(19,3))
go
insert into realtest (real1) Values (123456.123)
insert into floattest (float53) Values (123456.123)
insert into decimal9test (decimal9) Values (123456.123)
insert into decimal19test (decimal19) Values (123456.123)
go
insert into realtest (real1) select real1 -1. from realtest
insert into floattest (float53) select float53 -1. from floattest
insert into decimal9test (decimal9) select decimal9 -1. from decimal9test
insert into decimal19test (decimal19) select decimal19 -1. from decimal19test
go 21
go
select sizemb = SUM(sizemb), tablename from (
select SizeMb = (p.in_row_reserved_page_count*8.)/1024.,
tablename = o.name, indexname = i.name
from sys.dm_db_partition_stats p
inner join sys.objects o on p.object_id = o.object_id
inner join sys.indexes i on i.object_id = o.object_id and i.index_id = p.index_id
where o.is_ms_shipped = 0
) x
where tablename in ('floattest', 'realtest', 'decimal9test', 'decimal19test')
group by tablename
order by sizemb desc
go
select top 1 * from realtest
select top 1 * from floattest
select top 1 * from decimal9test
select top 1 * from decimal19test
go

First, the resultsets showing the storage of 2+ million rows in a single-column table.

tablenamerow_countsizemb
realtest209715226.382812
floattest209715234.445312
decimal9test209715228.382812
decimal19test209715236.507812

Here's the values that were actually stored, notice the important differences.

real1
123456.1
float53
123456.123
decimal9
123456.123
decimal19
123456.123

You can see that the ~10mb we saved with the real (or float(24)) data type isn't much good because of the catastrophic loss of precision. And even with 2 million rows, we've saved roughly 2mb of space by using float.

Now, let's run the same test, instead with the value 123456.1234567 instead. (The Decimal9 table has been eliminated from this result because it can only store a number with scale 9.)

create table realtest (real1 float(24) )
create table floattest(float53 float) --float(53) is the default
create table decimal19test(decimal19 decimal(19,7) )
...
tablenamerow_countsizemb
realtest209715226.382812
floattest209715234.445312
decimal19test209715236.507812

real1
123456.1
float53
123456.1234567
decimal19
123456.1234567

Note again that real is wholly unacceptable with its transparent loss of precision, while float and a properly-aligned decimal data type store the data appropriately and yet without a significant difference in storage (less than 6% difference).

Now, to slam the door on float.

Let's apply DATA_COMPRESSION = PAGE to this data.
create table realtest (real1 float(24) ) with (data_compression = page)
create table floattest(float53 float) with (data_compression = page)
create table decimal19test(decimal19 decimal(19,7) )with (data_compression = page)
...

tablenamerow_countsizemb
floattest209715226.382812
decimal19test209715224.320312

We can clearly that page compression reverses the small advantage that floattest enjoyed.

Since page data compression is a very popular, very widespread and very useful tool on modern Enterprise edition SQL Servers, we can rule out float as having any space saving advantage. In short, float loses the last ground it stood on.

The real and float data type columns only compressed by 15% and 23% respectively. The fixed decimal data type compressed, in this case, by 43.5% (36.5mb to 24.3mb).

If we scale the page compressed data comparison up to 134 million rows...

tablename
row_count
sizemb
floattest
134217728
1686.070312
decimal19test
134217728
1556.015625

...we confirm that floattest has lost its so-called space savings advantage, and is more than 7% larger than the same data in a proper decimal data type.

3/29/2014

South Alabama Day of .NET Presentations Downloads

I have been running into longtime SQL Saturday friend Ryan Duclos at SQL Saturday events in Baton Rouge and Pensacola since the beginning SQL Saturday days, so when he headed up a Day of .NET in Mobile, AL at the University of South Alabama, I was happy to join the agenda.

Big thanks and major kudos to Ryan (and wife) for organizing #SADoDN (or "Sad Odin" LOL). A job well done!  Also thanks to all attended and made it a great event!

I presented two sessions:

9:30am
SQL Admin Best Practices with DMV's [Beginner] Download slidedeck and sample scripts here

11am
SQL Server Permissions and Security Principals [Beginner] Download slidedeck and sample scripts here

3/27/2014

South Alabama Day of .NET 2014

About 200 miles east of Baton Rouge there lies a beachside town called Mobile, AL that will settle down long enough from spring break to host the first South Alabama Day of .NET at the University of South Alabama. The wife and I are looking forward to presenting at the event on Saturday, March 29, where we will both be giving back-to-back presentations in the morning.

In the Business Intelligence/Database track, look for:

  SQL Admin Best Practices with DMV's [Beginner] In this practical and script-focused session, we'll discuss many best practices regarding SQL Server administration, maintenance, optimizations and monitoring using Dynamic Management Views. DMV's are essential tools for the SQL administrator, but have a wide range of applications by developers, network engineers and DBAs. We will delve into best practices for server maintenance, index fragmentation, wait type analysis and future features. This session will be informative and eye-opening to anyone from a junior DBA on up.

  SQL Server Permissions and Security Principals [Beginner] A review of SQL Server permissions starting with the basics and moving into the security implications behinds stored procedures, views, database ownership, application connections, consolidated databases, application roles and more.

See you there!