So far in this series, I’ve talked about the advantages of VDI and Desktop-as-a-Service (DaaS) over traditional desktops and laptops, looked at the challenges that get in the way of VDI success, and examined how Nutanix HCI overcomes those challenges. In this blog, I want to focus on DaaS.
Designing, building, running, and maintaining VDI in-house can be challenging due to constraints that may include skills, datacenter space, budget limitations, and staffing. IT teams may already be oversubscribed. Finding and retaining experienced administrators with the skills to run a modern—but also complex—VDI platform can be difficult. Adding VDI to your existing mix of applications and services may just multiply your challenges.
Why choose DaaS
Here are some of the reasons you might choose Desktop as a Service (DaaS) for some or all of your needs:
-Accelerate delivery of virtual applications and desktops to users leveraging public cloud infrastructure in single or multiple cloud regions.
-Support hybrid cloud deployments with virtual applications and desktops running on-premises or in public clouds, all managed from a single control plane.
-Address the needs of temporary, seasonal, and contract workers
-Reduce capital costs(CapEx)/switch to a pay-as-you-go (OpEx) model
-Eliminate the need for VDI infrastructure deployment and management
Many companies also find that DaaS give them greater flexibility and agility to:
Support new business initiatives more quickly
“SaaSify” classic Windows applications, by enabling them to be run from a browser
Support a strategic transition to Windows 10 Cloud PC and run virtual applications and desktops closer to the data
Enable users to move to Chromebook and Google Enterprise and run Windows apps in a cloud-native fashion
Make VDI simple. Focus on applications and data, not the platform that supports their delivery
Before you can make an informed decision about DaaS, you’ll want to know more about it.
What is DaaS in cloud computing?
DaaS provides many of the same capabilities and advantages as VDI for end-user: anywhere access to applications and desktops, no data on the end-point, and centralized management. However, rather than design, procure, deploy, and manage all the necessary hardware and VDI software yourself, with DaaS you contract with a vendor who develops and operates the platform to satisfy your virtual desktop and application needs.
It’s an easy mistake to assume that once you choose your DaaS provider, there’s no work left for your team to do. However, in most cases that’s not completely correct. Your IT team or IT partner will still be responsible for configuring, supporting, and maintaining the user workspace, including the applications that your end-users will have access to. This is illustrated in the figure below. I’ll discuss the three layers of this figure from top to bottom.
Software-Defined Workspace Layer
I’m starting at the top, because the top layer is the one you’re directly responsible for. If you think about this for a moment, it makes sense that you have to be responsible for defining and managing the workspaces and the applications that your end-users will access, so you can ensure that they have access to the set of applications they need—whether that’s commercial software or in-house applications.
The virtual workspace includes the Windows user profile and application settings—all configured and managed centrally. Applications are installed and published within the workspace using central image management solutions. These applications often require connections to other back-end systems like web services, database services, and file services to function correctly, so integration between the workspace and back-end services from both an identity management and networking perspective is an important element of the solution.
Desktop-as-a-Service Layer
The next layer down contains the primary thing that differentiates VDI vs. DaaS: where the control plane, usually referred to as the broker, runs. The broker software, delivered as a service, handles the necessary access provisioning, security, and management functions in a virtual desktop or application environment. It helps to ensure that users are connected with the appropriate desktops and applications running on the appropriate infrastructure which can be in public clouds or on-premises.
The broker may connect with existing identity providers such as Okta, Google, or AzureAD, and it interacts with virtual infrastructure on-premises or in public clouds to power on/off, create/delete, and reboot the workload VMs that run end-user applications. The broker is also responsible for secure role-based access control and gateway services to deliver secure access to virtual applications and desktops via a launchpad, web interface, or APIs. With VDI, the broker and its supporting services runs in your datacenter (or co-location) and is entirely your responsibility. With DaaS, the broker is a managed service running in the cloud. A vendor, such as Nutanix, helps ensure that the broker and any associated services are up to date (often it is updated with new functionality weekly) and meet the agreed-upon SLAs.
Choose on-premises infrastructure:
-For predictable, consistent workloads
-To achieve fixed costs, pay for peak load 24x7
-When significant data is on-premises (for data locality)
-If there is no public cloud IaaS region near the end-users
-When no public cloud resources or capacity available
-If data must be on-premises for legal and/or regulatory reasons
Choose public cloud IaaS:
-For agility and fast access to resources
-To support variable, ‘bursty’, or seasonal workloads
to achieve a usage-based, pay as you go cost model
-When data is already in the cloud (data locality)
-To support a global workforce leveraging multiple cloud regions
-Don’t be a lemming; a cloud-first strategy doesn’t mean cloud-only. The DaaS broker should be able to support hybrid cloud. Leverage a combination of public cloud and on-premises infrastructure where they make the best sense to address business needs.
Does DaaS Address VDI Challenges?
Before I wrap up, let’s go back to the three VDI challenges from blog 2 to see how DaaS stacks up:
-User experience. Assuming you’ve done your homework to understand what your user needs are, the right DaaS solution should be able to meet those needs—and to a large extent it’s the provider who is responsible for day-to-day performance, not you.
-Complexity. The provider takes over the responsibility for operating and ensuring the availability of the broker, so any complexity becomes their problem. But, make sure you ask about the tools your team will use to configure desktops and applications.
-High or unpredictable costs. Modern DaaS offerings should give you a choice of what infrastructure to use to address your goals. If your goal is to move from CapEx to OpEx, DaaS + public cloud checks this box, and once you understand your DaaS provider’s pricing model, costs should be predictable. A hybrid deployment can reduce your total costs assuming you already own the necessary infrastructure. DaaS + on-premises infrastructure requires modern infrastructure to control costs.
To learn about Azure Virtual Machine Pricing visit, CloudDesktopOnline.
