VCAP7-DTM Design study guide - part 1

I recently passed my VCAP7-DTM Design (3V0-752) at VMware Empower 2019 in Lisbon and wanted to share my experiences and give you some advice on a small VCAP7-DTM Design study guide.
As I recently passed the VCP-DTM 2019 exam, I thought the VCAP exam would be similar but this was not the case. It was a difficult exam that tested all of my knowledge about Desktop and Mobility.

Note: This is a VCAP7-DTM Design study guide that will guide you in understanding the VMware Methodology in creating a Desktop & Mobility (Workspace ONE) design.
No answers or tricks for passing the exam will be given.

Index:

VCAP7-DTM Design study guide Part 1 – phase 1 & 2: logical design
VCAP7-DTM Design study guide Part 2 – phase 3: physical design
VCAP7-DTM Design study guide Part 3 – phase 4: physical design
VCAP7-DTM Design study guide Part 4 – phase 5: network design

Certification track:

First and foremost, let me give you some insight into what the VCAP7-DTM certificate exactly is.
The VMware Certified Advanced Professional is the advanced level certificate that is for those who design and build VMware solutions (VCAP Design) and manage and optimize VMware solutions (VCAP Deployment). The VCIX (Implementation Expert) is the combination of both VCAP’s.

Content and versions

When I started studying, the first thing I noticed was that the exam is based on older versions of the VMware products. This was already mentioned in the exam blueprint. A quick rundown of all versions from the blueprint are listed here below:

Horizon 7.2
UEM 9.1
IDM 2.8
AppVolumes 2.15
vSphere 6.5 with vSAN 6.2
Thinapp 5.1
vRops 6.4

Make sure that you use these versions when studying for your exam and know the limitations and changes that correspond with these.

Exam Sections

The deploy exam is designed to help you in making a solid design based on the VMware methodology. A solutions-driven approach will be used that will identify all high-level requirements and build your design based on the key requirements that have been identified. The blueprint mentioned above identifies the following sections that a candidate needs to know:

Section 1 – Create a Horizon Conceptual Design
Section 2 – Create a Horizon Logical Design
Section 3 – Create a Physical Design for vSphere and Horizon Components
Section 4 – Create a Physical Design for Horizon Storage
Section 5 – Create a Physical Design for Horizon Networking
Section 6 – Create a Physical Design for Horizon Desktops and Pools
Section 7 – Incorporate Application Services into a Horizon Physical Design
Section 8 – Incorporate Endpoints into a Horizon Design

I will be looking into every section individually and give you some essential tips and tricks that you need to have to succeed in the exam. I will also be linking some references to other great bloggers that have helped me in studying for the exam. They have spent the effort and time in creating similar in-depth VCAP7-DTM Design study guides, so I highly recommend reading them.

Kyran Brophy: http://www.euc-kiwi.com/vcap7-dtm/
Michael Rebmann: https://www.cloud13.ch/2018/08/16/vcap7-dtm-design-exam-part-1/

Section 1 – Conceptual Design

The conceptual design is one of the most important phases when starting on your design. It will identify all business drivers that will correspond in X amount of use cases. These use cases should be unique in the requirements they need. The requirements should be checked with the SMART framework. If they don’t meet these 5 criteria, I highly suggest to drop them.

A requirement can be functional (what the system should do) or non-functional (how the system performs a certain function).

After validating all requirements, list them from top to bottom in order of importance and impact on the design.

To make sure your conceptual design is covering all aspects of the use case requirements, list the possible risks, constraints and assumptions that could be relevant. These terms are typical in every conceptual design and are standard project management terms. Therefore I won’t go into the exact terminology of each item but instead will give a relevant example.

Example: A customer wants a high available environment spread across multiple data centers. So the requirement could be “availability of 99.99% is needed”.

Risk: a possible risk could be that one of the datacenters is located in a flood zone. Or the interconnect WLAN between datacenters has a 100ms+ latency.

Constraint: an example could be that the budget is $500.000 dollars. Or there is limited bandwidth available due to a customer’s Internet Service Provider (ISP).

Assumption: an assumption could be more like the datacenter has sufficient cooling and power to house the new hardware.

At the end of your Conceptual design, you should have multiple use-cases that are well defined and have no major overlap with other use-cases. Why do I say this? If let’s say 2 use-cases have multiple similar requirements, why should they not be combined together?

Section 2 – Logical Design

As mentioned in the blueprint, the second section is about creating a logical design based on the outcome of the conceptual design you have made in phase 1.

This logical design is a high-level drawing on “HOW” are we going to provide the “right” solution for a use case. To be honest, this phase is actually the most important one in the entire design procedure. If we make a bad decision on the “HOW”, you can imagine the impact it has on the “WHAT” phases of our design (phases 3 to 6). It can fully change needs and limitations.

To give an example: let’s say a customer will be implementing Horizon VDI, using Windows 10 as the operating system. What if their primary ERP application was a 32-bit Windows XP only compatible application! What does it do to our design? Well, chances are likely that the application won’t work on the OS. So a solution could be to use a sandboxing solution like VMware Thinapp.

With all the use case requirements, we can start defining the CRAMPS. CRAMPS stands for a collection of criteria that can be used when making design decisions.

Cost
Recoverability
Availability
Manageability
Performance
Security

When looking at these from a “Workspace ONE” perspective, each decision can have a massive impact on the rest of the design. Therefore it is essential that you have a good knowledge of the following Horizon principles and always link them with the following questions:

How does it influence CRAMPS?
Are there limitations when using this?
Does the solution have any prerequisites?

To give you a better understanding of what I mean, I will give some examples. From a design perspective, phase 2 is the most essential part of your design. This needs to be right or all the following phases will be building on an incorrect logical design. Therefore think twice and use the questions mentioned above.

Example 1:

In a hypothetical scenario, we have a healthcare customer with around 3500 users working in a 24/7 environment. The environment needs to be high-available due to business requirements. The customer wants to provision an RDSH farm to host all users Desktops.

So, with this example, we can start thinking about “HOW” do we provide the best solution for this use-case.
Let’s say, we want to use a Horizon cloud pod architecture (CPA) with the use of automated Instant Clone RDSH Farms.

How does it influence CRAMPS?
Cost: CPA will drive the cost up as more building blocks are required. Resource and management blocks containing vSphere clusters are needed.
Recoverability: As the Cloud pod architecture has been designed from a resiliency and scalability standpoint, recoverability will benefit from the multiple blocks and or pods in the CPA.
Availability: Similar to recoverability, a cloud pod could consist out of more than 1 pod or multiple resource blocks in a pod, adding higher levels of availability.
Manageability: With the introduction of CPA , the management will be slightly more complex as more elements are added. Meaning more components need to be maintained and updated.
Performance: As CPA is highly scalable, the performance can grow and be adjusted to the business needs.
Security: With the introduction of CPA, more components are added to the environment which need to be maintained.
Are there limitations when using this?
The use of Cloud Pod Architecture will drive the cost up as more components are likely needed. If in phase 1, the budget was mentioned as a primary criteria for this use-case, CPA could not be the right decision.
Does the solution have any prerequisites?
No, but if we adjust the use-case a little, So it would state that for example, the customer has 2 data centers and they want to use this in an active-active setup. This change implicates that a global traffic manager (GTM) like an F5 is required. It could influence the overall design decision and has an impact on the budget.

Example 2:

In a second hypothetical scenario, a logistical enterprise wants a new digital workplace using VMware Horizon. Both x86/x64 applications and multiple desktops will be provided to the entire workforce. An in-place Single Sign-On (SSO) has recently been bought and must be integrated into the new solution.

So, let’s begin again and “HOW” do we provide the best solution for this use-case.
In this example, we will be deploying Identity Manager (IDM) to be integrated with the existing SSO solution and link it with the new Horizon environment.

So in this case, we have 2 options: deploy IDM on-prem or SAAS IDM (cloud-based).

How does it influence CRAMPS?
Cost: With the introduction of IDM, no major additional costs can be expected when deploying on-prem. With cloud-based IDM, an additional licensing cost does apply.
Recoverability: In the case of an on-prem deployment: the Identity Manager appliances will be deployed as a 3 node production cluster per Datacenter. When failing over some manual steps are required. Therefore implementing IDM has a negative impact on the recoverability of the environment.
Availability: In case of the SAAS Tennant, the availability number will be much higher than when deploying it on-prem. A 99.99% availability means that there is only a yearly downtime allowed of less than 53 minutes.
Manageability: When using the SAAS based IDM deployment, the management overhead will be much lower than when using a redundant multi-datacenter IDM deployment.
Performance: Performance could be impacted by a SAAS IDM deployment when the internet breakout is for example fully saturated.
Security: If security is a major requirement for the customer, going the SAAS method, could not be possible.
Are there limitations when using this?
When using identity Manager on-prem, the complexity of the environment does increase.
Does the solution have any prerequisites?
The on-prem deployment of IDM requires an always-on SQL cluster. This does also requires the necessary licenses.