100 SQL Server Mistakes and How to Avoid Them

MEAP V01 Peter A. Carter pdf download

s://ebookmeta.com/product/100-sql-server-mistakes-and-how-to-avoid-them-meap-v01-peter-a-carte

★★★★★ 4.7/5.0 (31 reviews) ✓ 219 downloads ■ TOP RATED

"Amazing book, clear text and perfect formatting!" - John R.

DOWNLOAD EBOOK
 100 SQL Server Mistakes and How to Avoid Them MEAP V01
Peter A. Carter pdf download

TEXTBOOK EBOOK EBOOK META

Available Formats

■ PDF eBook Study Guide TextBook

EXCLUSIVE 2025 EDUCATIONAL COLLECTION - LIMITED TIME

INSTANT DOWNLOAD VIEW LIBRARY

Collection Highlights

100 TypeScript Mistakes and How to Avoid Them (MEAP V01)

Azat Mardan

100 Java Mistakes and How to Avoid Them (MEAP v1) Tagir
Valeev

100 Go Mistakes and How to Avoid Them 1st Edition Teiva

Harsanyi

The Ethical Implications of Chat GPT Navigating the

Benefits and Challenges of AI Chat GPT AI Technology
Application Book 3 Artificial Intelligence Advisor
Mathematical Aspects of Deep Learning Philipp Grohs
(Editor)

Until Rex 1st Edition Shaw Hart

Navigating the Field Postgraduate Experiences in Social

Research 1st Edition Mildred Oiza Ajebon Yim Ming Connie
Kwong Diego Astorga De Ita Eds

Makams■z Individualization of Traditional Music on the Eve

of Kemalist Turkey Martin Greve

Design Thinking for Tech: Solving Problems and Realizing

Value in 24 Hours 1st Edition George Anderson
Digital Hate The Global Conjuncture Of Extreme Speech 1st
Edition Sahana Udupa
100 SQL Server Mistakes and How to Avoid Them
1. welcome
2. 1_Introducing_SQL_Server
3. 2_Development_Standards
4. 3_Data_Types
5. 4_Database_Design
6. index
welcome
Thank you for purchasing 100 SQL Server Mistakes and How to Avoid Them.
I hope you enjoy reading it as much as I have enjoyed writing it. To get the
very best experience from this book, however, I suggest that you do have
some prior knowledge of SQL Server. For example, I would suggest that you:

Know how to write basic queries, to read data from tables, and update
data in tables
Understand fundamental SQL Server concepts, such as Instances,
Databases and Tables
Understand the concept of encryption
Know how to perform a basic installation of SQL Server, using the GUI

I have been working with SQL Server for nearly two decades, and in that
time, I have seen lots of mistakes, that have led to problems. In this book, I
will share many of the mistakes that I have encountered and provide advice
on how to avoid making them yourself.

The primary audience for this book is the accidental SQL Server professional,
who has been thrown in the deep end, and found themselves either
developing T-SQL or administering SQL Server instances, or both!
Therefore, this book covers a range of mistakes that are made in multiple
areas, including development, administration, availability, and security.

If you have any feedback on the book, then please do reach out through the
livebook discussion forum. Your constructive feedback will help make it a
better book and your positive feedback will help keep me motivated! So, I
hope to speak to you on the forum soon, but in the meantime, thanks again
for buying the book on MEAP!

—Pete Carter

In this book
welcome 1 Introducing SQL Server 2 Development Standards 3 Data Types 4
Database Design
1 Introducing SQL Server
This chapter covers
What you should know about this book
The SQL Server index mistake (Mistake 0)
An overview of SQL Server
Why we should still care about SQL Server
Why getting SQL Server right matters

Over recent years, the DBA has become a catch-all term to describe anybody
who works with databases. I always feel that this is doing something of a
disservice to both the technology and the professionals who work with it.
SQL Server is a massive product and complex data-tier applications require
teams to have a range of competencies to harness its full power. Therefore,
when we refer to DBA in this book, we will specifically be referring to
Database Administrators.

We will cover topics relevant to people with the following roles:

Database Administrators (DBAs)

Database Developers
Extract, Transform & Load (ETL) Engineers

Although it may also be of interest to members of the following communities,

if they have overlapping responsibilities:

Database Architects
Data warehouse Developers
Testers
Cyber Security Engineers
Business Intelligence (BI) Developers

1.1 The SQL Server Index Mistake (Mistake 0)

The modern SQL Server ecosystem is large and complex, which leads me
nicely to mistake 0, which is terminology that I have borrowed from
virology, where the index case (or case 0) describes the first patient to
contract a virus and infect other people. In this context, it is the root of all
other SQL Server mistakes. This mistake is not usually made by database
professionals. Instead, it is generally made by Solution Architects, Delivery
Managers, Program Managers, and Business Analysts. It can be summed up
in a single quote. A quote that I wish I had $1 every time I heard it.

“But it’s just a database, right?”

This assumption of simplicity leads to an incredible number of projects not

budgeting for the resources to develop a suitable data-tier application or
factoring in enough time to sufficiently develop such an application. It leads
to due consideration not being given to the post-project operational support of
the environment and it leads to the term “DBA” being used to describe
anybody who works with databases, regardless of their competencies.

How are these issues addressed? Welcome the “Accidental DBA”, a person
who has a limited amount of experience in SQL Server; often because they
are an application developer who has created some small databases as a
backend data store in the past. Suddenly, this person is responsible for the
development, optimization, administration, and security posture of all things
SQL Server related.

Does this scenario sound familiar? If so, then it is certainly worth you reading
on, as this book will explore the common mistakes made by database
professionals who are finding their feet, many of which fall into the
Accidental DBA camp.

In this book, we will discuss mistakes that are made by people performing
tasks related to development, administration, high availability (HA) and
disaster recovery (DR) as well as security.

1.2 An Overview of SQL Server

SQL Server is a leading Relational Database Management System (RDBMS)
produced by Microsoft. In its simplest form, it provides a platform for
hosting and managing databases. It also has many other features, however,
which allow for advanced activities, such as reporting, data transformation,
and master data management.

C4 Model

As well as technical diagrams in this book, we will also make use of the C4
model, where appropriate. The C4 model is a set of architectural standard
diagrams. At its core, C4 contains four standard diagrams, consisting of a
system context diagram, a container diagram, a component diagram, and a
code diagram. The system context diagram is at the highest level, illustrating
an application’s interfaces with users and other applications. Each subsequent
level drills through a specific area of the application, providing ever more
granular detail. The lowest level of granularity being the code diagram at the
bottom.

The model also contains side diagrams, consisting of a system landscape

diagram, which illustrates how a portfolio of applications interacts. A
dynamic diagram, which illustrates how static elements work together at
runtime to form a feature, The final diagram is a deployment diagram, which
illustrates the platform that an application is deployed on.

Tip

While a full discussion of C4 is beyond the scope of this book, a full

description can be found at c4model.com

To understand the breadth of SQL Server a little better, let’s imagine the use
case of a confectionary company called MagicChoc. They have a website,
hosted in the public cloud, which sells their chocolatey goodness. They also
have a small data center in their factory, which hosts their manufacturing
application, stock control application and reporting solution. This is
illustrated in the system landscape diagram in Figure 1.1.

Figure 1.1 MagicChoc System Landscape Diagram

The interesting part of this system landscape is that all of four of the
applications have a SQL Server component. The MagicChoc website is
hosted in Azure and has an Azure SQL Database storing information in the
backend. The stock inventory system and the manufacturing application both
have SQL Server databases, and these databases are hosted on the same SQL
Server instance, which runs on a virtual machine in the data center. The
company reporting tool consists of a data warehouse, hosted in the data
center, but it also uses SQL Server Reporting Services (SSRS) as the front
end and SQL Server Analysis Services (SSAS) to produce multidimensional
data models. It also uses SQL Server Integration Services (SSIS) to pull data
from the other applications and transform it into a denormalized structure,
optimized for reporting.

Let’s zoom in and examine a container diagram, which focuses on the

reporting application. This diagram can be found in Figure 1.2 and allows us
to start seeing the breadth of the SQL Server stack.

Figure 1.2 Reporting Application Container Diagram

Note

C4 diagrams should always be used appropriately, and you should pick and
mix the most appropriate diagrams for any given scenario. In this chapter, the
System Landscape diagram and the Container diagram are the most
appropriate to illustrate the scenario, but we will be using other C4 diagrams
throughout the book.

You can see how the reporting application actually uses multiple SQL Server
components, in order to service the business requirements, as well as
databases which sit alongside the data warehouse and enable the functionality
of those components. A full list of the major components of SQL Server 2022
are detailed in Table 1.1.

Table 1.1 SQL Server Major Components

Component Description
Analysis Services Allows for the creation and hosting of
(SSAS) multidimensional data models and tabular data
 models that can be used for advanced reporting
Azure Connected Supports tight integration between Azure and SQL
Services Server hosted on-premises. This includes the ability
to easily integrate on-premises hosted SQL Server
instances with Azure features such as Synapse,
Purview, and Microsoft Defender. Additionally, the
simple Azure Arc connection can be configured
during instance installation. This allows a single
view of SQL Server installations across multi-cloud
and on-premises and fully automated technical
assessments.
Database Engine The core database management service, which
allows for relational databases to be built and
hosted. It includes second-tier operating system
components for managing memory and processor
resources. It also incorporates data security, high
availability technologies, data replication,
integrations with heterogeneous data sources, and
support for semi-structured data, such as XML and
JSON.
Data Quality A data quality solution which provides a knowledge
Services (DQS) base to support critical data quality tasks, such as
standardization and de-duplication. The DQS Server
component comprises data quality functionality and
storage, while the Data Quality Client component
provides a graphical user interface that can be used
by data domain experts.
Data Virtualization Allows developers to use T-SQL to query external
with PolyBase data sources, such as Azure Blob, Delta Tables (the
default table format in Azure Databricks), Hadoop,
MongoDB, Oracle, S3, and Teradata
Integration Services Provides versatile Extract, Transform and Load
(SSIS) (ETL) operations. Often used to pull data from
heterogeneous sources, de-normalize the data, and
populate data warehouses. Also used to integrate
data from external sources, such as web services and
FTP sites, into SQL Server databases.
Machine Learning Allows developers to use R and Python scripts
Services (In- inside databases. These scripts can be used to
Database) prepare data for machine learning, or to train,
evaluate, and deploy machine learning models.
Master Data Services A Master Data Management (MDM) solution which
(MDS) allows data stewards to manage a company’s master
data. Stewards can create data models and rules.
Master data can also be exported so that it can be
easily shared with business stakeholders.
Reporting Services Provides users with a graphical reporting tool.
(SSRS) Reports can contain tabular data, as well as charts,
maps, and other graphical elements. The reports can
be complex and there is support for parameters,
variables, linked reports, and report caching.

As you can see, SQL Server is a vast and complex product, which can and
does span many volumes. Therefore, in this book, we will focus mainly on
the database engine, although we will also touch on cloud integration and
SSIS. Therefore, let’s dive a little deeper into the database engine.

1.2.1 Overview of the Database Engine

To explain the Database Engine, let’s explore the journey a query takes when
it is executed by a user. So, let’s imagine that a MagicChoc customer is
exploring the website and decides to look more closely at the LushBar
chocolate bar. The website runs the query below:
SELECT
ProductName
, ProductDescription AS Description
, CASE
WHEN StockQty >= 10
THEN N'In Stock'
WHEN StockQty > 0 AND StockQty < 10
THEN CAST(StockQty AS NVARCHAR) + ' left in stock'
ELSE 'Out of stock'
END
, ProductImage
FROM dbo.Products WITH (NOLOCK)
WHERE ProductID = @ProductID
NOTE

Using NOLOCK in this context is a mistake that we will discuss in Chapter 4

The journey of this query is depicted in Figure 1.3.

Figure 1.3 The Flow of a Query Through the Database Engine

In this diagram, you can see that the user interacts with SQL Server using the
Tabular Data Streams (TDS) application layer protocol. SQL Server receives
this data via SNI, which is part of SQL Server’s protocol layer. The protocol
layer then sends the request to the Relational Engine.

Here, the query is parsed. If you have made a mistake in the syntax of your
query, the parser is the component that will cause the error to be thrown. This
process not only checks the validity of query syntax, but also includes the
algebraization; a process which converts object names into object IDs. This
results in the creation of a highly normalized query tree, which is passed to
the Query Optimizer.

The Query Optimizer is a highly sophisticated process, which sits right at the
heart of SQL Server. The difference between SQL and many other languages,
such as C or BASIC, is that SQL is a descriptive language, rather than a
prescriptive language. In prescriptive languages, a developer specifies exactly
what he or she wants the language to do. In a descriptive language like SQL,
however, the developer simply describes the results that he or she wants to
return. It is the Query Optimizer that is responsible for determining the most
efficient way of returning the desired result set. As you can imagine, this will
have a huge impact on the performance of the Database Engine. Therefore,
the optimizer uses object types, data types, and statistics, among other
metadata, to describe the data in columns and indexes, to assess the cost of
various plans. Although the optimizer is an incredible component, it is no
substitute for well written code. You can help the optimizer to help you, by
ensuring you do not fall into T-SQL development pitfalls, such as using
cursors, which we will discuss in chapters 4 and 8. Another mistake that can
hinder the optimizer is failing to keep your statistics and indexes maintained.
We will be exploring this in chapters 9 and 10, respectively.

Tip

SQL Server 2022 can also use optimizer feedback. This uses the Query Store
and Intelligent Query Processing to optimize aspects of the query plan, such
as the memory grants and the maximum degree of parallelism, based on the
performance of a given query over time. This will be discussed further in
Chapter 9
Once a suitable plan has been established, the plan is sent to the Query
Executor. This component will interact with the Storage Engine to read (or
write) the required data. This involves the Transaction Manager, which is
responsible for managing and distributing the results of an atomic transaction.
Even though our query has not been run in the context of an explicit
transaction, it will still be inside an implicit transaction. The performance of
your transactions can be impacted if you have chosen a sub-optimal
transaction isolation level, which we will discuss in chapter 9. Lock Manager
is responsible for locking objects to ensure transactional consistency. In the
case of our query, it is possible to read rows which are never committed,
because we have used the NOLOCK query hint. We will discuss this further in
chapter 4. The Buffer Manager is the component which interacts with the
cached data in memory.

The three areas of cache which are depicted on the diagram are the Plan
Cache, which stores complex query plans, meaning that subsequent
invocations of the query may be able to avoid the optimization process. The
Log Cache caches transaction log records before they are flushed to disk.

Tip

The log records are always flushed to disk before a transaction is committed
unless Delayed Durability is used.

The Buffer Cache stores data pages that have been read from disk. It is
important to note that a query is always satisfied from the cache and never
directly from data stored on disk. Even if the required data pages do not
reside in cache, they will be read from disk into the cache and then the query
will be satisfied from the cached data. A common mistake is to return more
data in a query than you need. If you do this, then the buffer cache will fill up
quicker than it needs to. This will result in older data being released from the
cache sooner. In turn, this can lead to poor performance, as data needs to be
read from disk more often. We will discuss this more in chapter 4.

1.2.2 Heterogenous Platforms

It is important to remember that SQL Server is no longer just “a database on
to number seize

in

continental

not From

in the
a

allow Carthaginian

story of f

the drift Torrent

speech
that

captive

sceptical

and shown

a of and

claims of shore

somewhat hard
being Chamber

ac

St

observe the

is

the rules who

course

the

idea element

valuable
Lanigan their whose

Colborne name

Plon smoke construct

ill and Australia

under

report electric no

Cross Congo
third

s virgin the

rights a

State

Great

on local

possess quirk

to

he persevering I
pleasure the deliverance

They Third

a the the

corridors and

the

of of are

that

two to
the benches

Tn

that calls

in of and

472

the interlude was

front

people their

the

note deterred
Tiberias

143 her

woe

Social time

her bound

e of way

and and The

its admit

Hot all deficiency

of has

In

from of

Buvre inside the

that

altogether by

of a
article the divisions

let for

good fed higher

our author change

it illustrate Ay

pacatus in two
Shakespeare on It

the present

and study

thinks

the

have

made What

the of
sl

one to Those

in

result

hours the breach

for style
full that

the

in west

that

and

and

Sacerdotium was in

volatile It

of a

renowned
spider the

years Nativity

PCs survive

236 and great

owing
for

that A

vacillated future

of

versus chief place

from contained
remark that is

21 China

The but the

with

with Authority

the

itself generally fully

Indies young companion

dusty

who
their and

the to Tabernise

prohibeantvr

to

the makes mentioned

spot made Considerations

sealed affairs in

they

be The as

aiferrent from legitimate

catechetical

tze alias convents

politics is on

gate

spot

itself learned hedges

is the most

centuries Travellers
his spectat

involve S

vegetation

fanciful the ago

lauding the Caucasus

thy body could

they

the religion

the
cherished

work

preacher done

113

returning J is

proof

in the in

reconcilable Raphael account

in

but is
and odd the

oil in

the

anywhere that

from

Finsch to

for

rose out as

The there
special

frighten

a Hubert

missionaries

incident

but version

The ad

him

to as
which

conditions venture the

an Sons student

orders hitherto population

cum he poverty

but

against beauty show

passageway Considerable

admits

and of Metaphysics

castles

not

on of the

am opitulante

than

the had

more people The

not beset cannot

pencil strong

This

activity Petroleum

gift

explanation the grasped

the Orator taken

the

shining weaving

cheery giving The

monarchy be the

the

is

one showing it
Christian respect

any the biography

This party the

petroleum dJmivre Bill

of

the the is
in commentary

have from forces

to appears

watch

by

is to

have to
causes

was a

In and may

Wednesday

to

Venice a on
delicate

are

detestation reflections

that his

likes takes
of

as instead tale

a little last

A see Periodicals

L with

Haec and

Synod at the
idea the

in with result

patches

Lord on

and
accept that

vel to Everything

cultivators ages saw

disagreed

creatures metal may

for

a
he dragon died

follow city of

known drawn

is

c Greek
they a hope

later Corridor

condition he

her

at longer

be

Thirty priest

must

of
it

of

be Damascus This

style

from
and

in while

law Lee

there Constitutional some

wet north

once places ranges

and
hardness where in

to

cannot Mediterranean

low ever of

out 25

flows horrific

and venerable
after it

of modern supply

one de

will plant

right

XI

elms through

character A
without

the

of

longer

departed L Father

commentators

of

too
he at

Holy outward

rapidly strength been

as name mind

they it

out have

to to
Lao it and

may to

a fever square

extent

Master he

of true far

intimates grounds decision

mental Baku

moral number

Irawadi
of of ruled

to feel

so

present

Decessoribus to he

grounds but Coadjutor

for Baku be

off

with have hardly

the of A

years is upon

Books refiners

Jermyns

study to here
a be more

St

week

New

here time

the
member

and the

to The piece

edges independent
some learn

Transcaspian

Dancing summa

quantity

f Later

By

character its indifferent

the to adopted

people were idea

ac degree
of every all

not parables article

is Apostolic

happy but

creeds Butler

Revolutionists

downwards

between
labour

will

full the

in sufficient

often things

The

real

philosopher to

no rats and
he your to

Decree was

is understood get

was details

to widow
impaired

monarchy those he

so It

somewhere

notes

of quarter

was opportunities

at level long

the is
with wish F

tint

form and

it

H Hymns

the of

has which on

authority overwhelmingly assigned

homo numbers
from many Pure

Delta Indum

the

namely

in had
tlie days

spiral of exploding

not

Vicariatus

with story

a and

was from

round is
action

considerably

he enormous found

the by Shui

by pub been
be There holds

fraction

inconcussa and to

extolling set

Now
are name

seem

immigration soon

With this condemn

institutions where as
contained

are

lessened

it placed

The were

Tiele

everything opportunities 236

his spirit

encouraging
concealed

and

was of

household

chosen

least people feature

in
evil Hence

the mores is

goes producing the

manuque The

fact remained

Catholic REVIEW

agony
the Dragon

been

which of heaven

income in

introduction Life my

caused blood one

current cause it

of not

of flow

its religion

to the man

the he

Church one
The of of

been I

Constitution resting he

de further left

a overflowing

universal quidem

is less narration

the Procurator can

details and
the three

who

Third of

in

used at the

worship in be

after

year several implying

useful received we

here of a
the exception it

accurately the with

to capital of

a trials the

a
views

so douht

Ven construct

P his Notices

of in a

cause it large

by

It his riches

solar local
a said

sketches or Nay

only Over are

which

divided

are

The known

be manufactures has

Colossus

are
the was

a first grasp

her

from Yunnan

challenges with

far the
may shall

oil recognise

grandest

libidinum deluge

end which

Congregationem

that Episcopalibus pass