Author: Prashanth Jayaram et al
Publisher: Independent
Pages: 622
ISBN: 979-8706128029
Print: B08Y4LBTP4
Kindle: B08XZQJHMK
Audience: Azure DBAs
Rating: 2 or 4 (see review for details)
Reviewer: Ian Stirk

This book aims to help you pass the Azure Relational Database exam DP-300, how does it fare?

Businesses are increasingly aware of the advantages of cloud-based systems (e.g. cost, scalability), this has resulted in a growing interest in, and migration of systems to the cloud. This also means a corresponding increase in cloud-based DBA jobs.

Microsoft provides popular cloud-based database options, called Azure. If you want to increase your job prospects, it makes sense to learn about Azure. One way to do this is via certification, and this book covers the syllabus for the Administering Relational Databases on Microsoft Azure exam (DP-300).

This book is aimed at DBAs and developers who want to study for the exam. Some prior knowledge of on-premise SQL Server is assumed.

Below is a chapter-by-chapter exploration of the topics covered.

Chapter 1. Azure Fundamentals and Concepts

The book opens with details on the growing demand for cloud databases, providing reassurance that this is the correct direction for your career. In particular, taking the Administering Relational Databases on Microsoft Azure exam, should give you an advantage. Most of the chapter then gives some background basics, which provide the base for the subsequent chapters. 

The different cloud models (public, private, and hybrid) are outlined. Next, the different service types (IaaS – Infrastructure as a service, PaaS – Platform as a service, and SaaS – Software as a service) are examined, the underlying difference is the tradeoff in the amount of control you have over the server versus the increasing number of services provided automatically by Microsoft (e.g. backups). There’s a helpful discussion concerning architecture, in relation to regions, geographies, paired regions, availability sets (update domains and fault domains), availability zones, and resource groups. Next, Azure networking services are outlined, discussing a virtual network, load balancer, VPN, CDN. As is often with Azure books, I find too much knowledge of networks is assumed. The chapter ends with a discussion of the various Azure Storage services, namely: disk, blob, files, queue, and table.

The chapter provides wide coverage of the exam syllabus topics, providing useful diagrams and discussions. The vocabulary used is often good, even some colloquialisms, however, the grammar is sometimes bad, making it difficult to understand the sentences (e.g. “This chapter will learn about Azure SQL deployment options to bring value to add the business”). While it is possible to understand the book, more so if you already have a background in SQL Server/Azure, it shouldn’t be this difficult. Since there is bad grammar in every chapter, this book needs an English editor! These traits apply to the whole book.

Chapter 2. Explore Azure Data Platform Roles

The various Azure data roles are outlines here, specifically: 

• Azure data engineer 

• Azure database administrator 

• Azure data analyst

• Azure data scientist

• Azure AI engineer 

There’s a helpful section on the on-premise DBA, and how they might move to become an Azure DBA. I suspect this is currently the most popular pathway, since perhaps 85% on your on-premise SQL Server knowledge is applicable to the Azure database environment. 

Equally helpful, there’s a section on migrating to the cloud, covering topics such as: education, migration tools, baseline metrics, security, and use of small migration projects to build confidence. Businesses are concerned with costs, so you may need to explain the advantages of moving to the cloud. The importance of TCO (Total Cost of Ownership) and ROI (Return on Investment) is explained. There’s a useful link to TCO calculator.

Overall, a useful chapter, it discusses some of the roles on the Azure platform, together with helpful advice on how you can support your proposal to move some of your business to the cloud.

Chapter 3. Azure Data Platform Choices in Azure

Here the various categories of data are discussed, namely: structured (relational data, rows/cols), semi-structured (NoSQL, no fixed schema), and unstructured (e.g. flat files, images, PDF docs).

Next, we look at the Azure database services, covering both relational and non-relational databases, and examining the degree of control you have on IaaS (i.e. Azure SQL on VM) and PaaS (e.g. Azure Database). Provisioning the databases via the Azure Portal or ARM templates is discussed – the latter is better for multiple deployments. 

The various database services are discussed: 

• SQL Server on Azure VM: There’s a useful list of questions to ask when deciding to use SQL Server on Azure VM (e.g. compatibility issues).

• Azure SQL database: allows focus on design and performance, with less need to worry about infrastructure (e.g. software patching). 

• Azure Managed Instance: used to migrate quickly and easily from on-premise to the cloud, especially useful when you want to use existing on-premise functionality that’s missing from Azure SQL Database (e.g. cross database queries, SQL Agent). 

• Several Open-Source databases exist on the Azure SQL platform, including: MySQL, MariaDB, and PostgresSQL. Their setup and capabilities are discussed.

• Azure Synapse Analytics: This is Azure’s Data Warehousing solution, although it goes beyond traditional data warehouse functionality, 

• Azure Cosmos database: non-relational multi-model database service (i.e. supports many types of NoSQL database). very scalable, stores data across azure regions (for improved performance). Supports MongoDB, Cassandra, Tables, etc.  

There’s a useful image summarizing the major differences between Azure SQL VM, Azure Database, and Azure Managed Instance.

This was a good core chapter, wide-ranging, with a useful examination of the various database offerings. I did feel the topic of non-relational databases was a bit weak (when compared to what was expected in the exam).

Chapter 4. Plan and Implement Data Platforms Resources in Azure

Often a key decision involves choosing between using either IaaS (Azure VM) or PaaS (Azure SQL). There’s a useful table of their differences that can aid you in this selection. With IaaS you have more control over your server (e.g. patching), whereas with PaaS there are many automatic admin features included (e.g. backups). Typically, you will use IaaS if you need to use legacy features that aren’t included in PaaS (e.g. access to the OS). 

There’s a useful section on deploying SQL Server on Azure VM, covering the various license types (PAYG, BYOL, AHB). There’s a brief look at the various machines in the VM family, that have various optimizations (e.g. optimize for CPU or Memory). Lastly, there are step-by-step walkthroughs for provisioning and deploying SQL Server to Azure VM using via Azure market place, Azure PowerShell, Azure CLI, and ARM templates. Next, there’s a look at migrating on-premise SQL Server to Azure VM. Various steps are outlined: discover, assess, migrate, cutover, and optimize. The chapter ends with a look at PaaS options for deploying SQL Server in Azure, covering: pricing model (DTUs and vCores), Hyperscale (up to 100TB, highest performance, scale up/down), Serverless (auto scale up/down, auto pause), and Elastic pools (share resources between databases). SQL Managed Instance is more akin to on-premise SQL Server. There’s a useful table comparing the features of Azure SQL Database and Azure Managed Instance. Finally, there’s a brief look at deploying MariaDB, MySQL, PostgreSQL on Azure.

This chapter has some very useful lab exercises, in which you’ll gain some practical experience. There is a useful calculation that can be used to determine if databases should be moved to an elastic pool, however this was not described here. The authors say you can use the DTU pricing model with Azure Managed Instance, however it is only applicable to Azure SQL Database. 

There’s a degree of overlap with some earlier chapters.