Part 1: The Exam

Intro

This is the first article in a three-part series about Microsoft exam AZ-103. In this post, we’ll briefly go over the history of this exam, take a quick look at its requirements, and then go over the exam’s structure.

Background

Azure is Microsoft’s cloud-computing platform. An Azure administrator is an Information Technology specialist who sets up and manages a company’s resources hosted in Azure. In the past, Microsoft certification exams were technology-oriented. For example, the exam closest to the job of an Azure administrator was 70-533: Implementing Microsoft Azure Infrastructure Solutions. In September of 2018, Microsoft revamped its certification programs. The new ones are supposed to be role-based. Exam 70-533 was replaced with the following two new exams:

AZ-100: Microsoft Azure Infrastructure and Deployment

AZ-101: Microsoft Azure Integration and Security

However, something did not click properly into place in the Microsoft universe. Maybe someone at the top decided that those titles were not sufficiently role-based or that having to pass two exams in order to achieve a single certification was unfair. A mere few months later those two exams were hastily retired and as of May 1, 2019, replaced with a shiny, “new & improved” exam AZ-103: Microsoft Azure Administrator. Contents-wise this new exam is the same as AZ-100 combined with about one half of AZ-101.

Requirements

Exam AZ-103 is very demanding, and passing it requires proper preparation. Unlike some past exams, this one is practically impossible to pass without having good understanding of Azure basics, which in turn requires a solid foundation in the related technologies. Before you even begin thinking about taking AZ-103, you should make sure you are well-familiar with the traditional basics, like TCP/IP networking, VPN, DNS, DHCP, Hyper-V, Active Directory Domain Services (AD DS), as well as some other technologies that are commonly used on on-premises corporate networks.

The exam’s requirements are organized into the following five sections:

• Manage Azure subscriptions and resources. – You are supposed to know how to create and manage Azure subscriptions and resources. You should know how to create resource groups, create resources in them, monitor those resources, and move them around. You need to be able to delegate authority for your resources.
• Implement and manage storage. – You should be familiar with the kinds of storage accounts and when each kind should be used. You should know how to create storage accounts, manipulate data in them, provide access to that data, and back it up and restore.
• Deploy and manage virtual machines (VMs). – You should be able to create, manage, and delete Azure VMs; back up and restore them; and migrate or replicate them by using Azure Site Recovery (ASR). You should know how to work with Azure Resource Manager (ARM) templates.
• Configure and manage virtual networks. – You should have thorough understanding of what VNets are and how they work. You should be able to create VNets and subnets; configure their IP address spaces; and implement routing as well as various types of connectivity, such as peering, VPN, and ExpressRoute. You should know how to protect VNets with network security groups. You should be able to deploy, configure, and troubleshoot certain types of network devices, such as load balancers and gateways.
• Manage identities. – You should know how to create and manage Azure users and groups, copy user accounts from on-premises AD DS to Azure Active Directory, and keep the two directory services synchronized. You must be able to implement different types of single sign-on (SSO) and multi-factor authentication (MFA).

Exam Structure

The number of questions might vary somewhat from one instance of the exam to another. Typically, you can expect about 50 questions plus two labs with approximately six to eight tasks in each lab. All those items are organized into half a dozen or so separate sections. You will have somewhere around three hours to answer all the questions and complete all the lab tasks. This exam is especially challenging because of the labs and the lack of separate time limits for individual sections. This arrangement makes it virtually impossible for you to review your answers and possibly correct some of them (except perhaps in the last section), even if you have plenty of time left. When you have answered all the questions in a section, you can’t afford to go back and double-check your answers. You need to keep moving forward because you have no way of knowing how difficult the remaining questions will be and how much time you will have to spend on the labs, which are known to be rather slow and often not to function properly. Once you exit from a section, you cannot come back to it until the entire exam is over. When it is, you’ll be given 15 additional minutes in case you want to go over the exam and write comments on individual questions. During the comment time, you will be able to see your answers, but you will not be able to change them. Your comments might help Microsoft to clean up some of its boo-boos but will in no way help you improve your score.

The order in which the sections are presented can vary. One of the sections might contain several series of Yes/No questions. The same scenario is repeated in three or four consecutive questions, and each question includes a different proposed solution. You need to answer Yes or No to whether the proposed solution helps you achieve the goal stated in the scenario. Typically, one of the solutions in each series is the right one and the remaining two or three are not. However, this is not guaranteed. Some series might include two or more correct solutions, and some series might have none. The main challenge in this section is that you are not allowed to return to already-answered questions. You have to evaluate each proposed solution only on its own merits, irrespective of any other proposed solutions. Sometimes, if in serious doubt and willing to take chances, consider answering No, because statistically, No is more likely to be correct than Yes.

There can be a couple of case studies with about four or five questions in each. Sometimes you might get only one case study and the other one might be replaced with a set of about four or five stand-alone questions in a separate section. The case descriptions are not too long or overly complicated, but the main challenge is not the technical substance but rather the case study format itself. Before you attempt to answer any questions, you have to read the entire case description at least once in order to become familiar with the situation. However, during the first read, you don’t know which particular pieces of information are significant and which ones are not. So, after reading the entire case description—it can take anywhere from a couple of minutes to up to 10 minutes, depending on how fast you can read—you start reading the questions, one at a time, and then you need to go back to the case description and sift through it again, trying to find those precious details that provide answers to the questions.

There will also be a section or two with regular stand-alone questions. And again, once you leave a section, you can’t go back to it.

The two labs can take a lot of time because they are usually slow, their tasks are not always stated clearly, and you are likely to encounter some technical difficulties. Put simply, the labs require a lot of time, do not always work properly, and might lead to your failing the exam despite your having adequate knowledge. If the labs indeed seriously misbehave on your exam, you will have two options. You can inform the exam administrator of the problem, subsequently file a complaint/appeal to Microsoft, and hope that Microsoft—after weeks or months of back-and-forth, behind-the-scenes deliberations, and attempts to dodge the issue—will eventually allow you to retake the exam for free. The other option is to cut your losses as early into the problematic lab as possible so that you will still have a fighting chance of salvaging the remaining questions in the remaining time and passing the exam. Well, on second thought, there is a third option: when going to the exam, keep your fingers crossed. Some people claim that it helps, too.

Conclusion

Despite some challenges, the exam is still passable. All you need to do is learn the necessary technical stuff and be smart about how you manage your time during the exam.

In the next post of this series, we’ll take a closer look at the labs and learn a trick or two that should increase your chances of success.

Read Part 2 here.