Virtual Desktop Infrastructure (VDI) technologies provide users with centralized virtual desktop environments. VDI is typically deployed on-premises by using single tenant models. Amazon Web Services (AWS) has created its own, cloud-based version of VDI, which is designed for multiple tenants. AWS VDI is a fully managed Desktop as a Service (DaaS) offering, called Amazon WorkSpaces.
In this post, we’ll explain the main differences between VDI and DaaS, explore use cases for Amazon WorkSpaces, and review best practices for Amazon WorkSpaces deployments. We will also show how NetApp Cloud Volumes ONTAP can help you efficiently manage AWS VDI resources.
In this article, you will learn:
Virtual desktop infrastructure (VDI) is the use of virtual machines to host desktop environments. These environments are hosted on a centralized server and can be remotely accessed by users as needed.
In VDI, a hypervisor is used to deploy multiple virtual machines (VMs) on a host server. A hypervisor is a piece of hardware, firmware, or software that creates and manages VMs. Each VM contains its own desktop environment, similar to a physical machine. When deployed, users can access these machines from any device via a connection broker. A connection broker is a software-based gateway that manages traffic between a user client and the host server.
When implementing VDI, you can create either persistent or nonpersistent desktops.
From a high-level perspective a VDI solution seems very similar to Desktop as a Service (DaaS) offerings but there are a few differences to be aware of.
Single tenant vs multi-tenant
VDI deployments use a single tenant model with each host server used by only one organization. Single tenant enables organizations to have full control over their server configurations.
DaaS has multiple tenants (organizations) on a single server. Each tenant has isolated access, but the overall resources of the host server are shared. In multi-tenant models, organizations do not have control over how a server is configured and are only allowed to manage their own provisioned resources.
Platform
VDI deployments are hosted on your own infrastructure and your IT team is responsible for configuring and maintaining servers and networks. Depending on the configuration of your networks, these deployments can be used without Internet access.
DaaS deployments outsource all of the server purchasing, maintenance, and management to a cloud vendor. These services also require Internet connectivity to access desktops. In exchange, DaaS offers scalability, automatic updates, and pay for use service.
Amazon WorkSpaces is a managed DaaS offering that you can use in AWS. It enables you to create and use Amazon Linux or Microsoft Windows desktops that are accessible from anywhere. With it, you can flexibly add and remove users while paying only for the resources you are actively using.
There are multiple use cases for Amazon WorkSpaces, including:
When adopting WorkSpaces, you are granted access to the free tier with limited resources for two months. Once you exceed those resources or time, you are charged either monthly or hourly.
Amazon WorkSpaces free tier
The free tier provides access to two Standard WorkSpaces with a 80GB root volume and 50GB for each of the user volumes. You can use these spaces for up to 40 hours total per month. The spaces are operated in AutoStop mode, which pauses the resource after a period of inactivity (1hr by default). Use time is counted while actively using the desktop up until the desktop is paused.
COVID-19 update
From April 1, 2020 through June 30, 2020, Amazon is offering the free tier with extended spaces. This offer is valid for all new WorkSpaces users, but the period ends June 30 regardless of when you start. It includes:
Pricing beyond the free tier
Amazon WorkSpaces offers desktop instances in three models:
Below we show Amazon WorkSpaces prices for the entry-level instances in each instance category, for the Windows bundles in the US East (N. Virginia) region. For up-to-date pricing and additional options, see the official pricing page.
Instance Type |
Specs |
Root Volume |
User Volume |
Monthly Pricing |
Hourly Pricing |
Value |
1 vCPU, 2 GiB Memory |
80 GB |
10 GB |
$25 |
$7.25/month + $0.22/hour |
Standard |
2 vCPU, 4 GiB Memory |
80 GB |
10 GB |
$33 |
$7.25/month + $0.30/hour |
Performance |
2 vCPU, 7.5 GiB Memory |
80 GB |
10 GB |
$45 |
$7.25/month + $0.47/hour |
Power |
4 vCPU, 16 GiB Memory |
80 GB |
10 GB |
$70 |
$7.25/month + $0.68/hour |
PowerPro |
8 vCPU, 32 GiB Memory |
80 GB |
10 GB |
$127 |
$7.25/month + $1.53/hour |
Graphics |
8 vCPU, 15 GiB Memory, 1 GPU, 4 GiB Video Memory |
100 GB |
100 GB |
$735 |
$22/month + $1.75/hour |
GraphicsPro |
16 vCPU, 122 GiB Memory, 1 GPU, 8 GiB Video Memory |
100 GB |
100 GB |
$999 |
$66/month + $11.62/hour |
When deploying WorkSpaces several best practices can help you ensure optimal performance and help reduce your costs.
Select the right bundle for your workload
You have a lot of flexibility when it comes to service bundles and payment in WorkSpaces. When choosing your options try to match your workload to the lowest functional option.
For example, if you are only performing basic tasks like email or document creation, you should be fine with 2GB of memory. Likewise, if you only need WorkSpaces for a few days or hours a month, you should choose hourly pricing.
VPC design
For greater security you can run WorkSpaces in an Amazon Virtual Private Cloud (VPC). This enables you to separate traffic from your external services and allows you to more carefully manage user access.
To use VPC you need to connect your WorkSpaces to Active Directory and define subnets for access. Once you set these subnets up, you cannot modify settings, so you need to double-check user groups and authentication before you finalize settings.
Consider zero clients
WorkSpaces enables you to use thin clients as endpoints. These clients are essentially screens with attached peripherals (mouse, keyboard, etc.). Thin clients do not contain an OS. Instead, a PCoIP chipset enables the client to accept transmitted data, like a WorkSpaces.
Because these devices don’t require all of the hardware of normal clients, power and space consumption is minimal. Also, although these clients are not cheaper than traditional clients, maintenance tends to be easier.
NetApp Cloud Volumes ONTAP, the leading enterprise-grade storage management solution, delivers secure, proven storage management services on AWS, Azure and Google Cloud. Cloud Volumes ONTAP supports up to a capacity of 368TB, and supports various use cases such as file services, databases, DevOps or any other enterprise workload, with a strong set of features including high availability, data protection, storage efficiencies, Kubernetes integration, and more.
To find out more about VDI deployment in the cloud and how Cloud Volumes ONTAP can help you run your VDI environment on cloud resources, download our guidebook on Virtual Desktop Infrastructure in the Cloud, where you’ll also learn about case studies of major companies who turned to Cloud Volumes ONTAP to make their VDI deployments cost-effective, highly available, and easy to orchestrate with the flagship NetApp cloud solution.