VDI Deployment Strategy for Healthcare
Overview
A full-platform strategy to maximize the utilization of a virtualized desktop infrastructure (VDI) in a hospital can present significant benefits for physicians and nurses to improve their workflow. Regardless of the benefits, an improperly managed VDI deployment can create significant disruption and introduce unforeseen problems which will overshadow the value of this technology.
Technology residing within a centralized data center, although easier to maintain and support, increases the potential risk of disruption in the event of a failure. For this reason, a well-planned VDI strategy will capitalize on the benefits this technology can provide while ensuring minimal disruption to clinicians caring for patients.
T2 Tech Group's experience has enabled the development of a strategy to minimize risk and disruption while simultaneously improving user experiences and ensuring a smooth implementation. A structured VDI deployment will include a thorough assessment of the environment prior to any design initiatives. Once the design is complete and built out, a comprehensive set of tests are needed to validate the functionality and resilience of the VDI platform.
VDI deployments can encompass several components. Within the full-platform virtualized environment, implementing Single Sign On (SSO) and secure printing solutions is recommended to improve efficiency and optimize clinical performance. Bundling the implementation of VDI, SSO, and secure printing can be more efficient than implementing them piece-by-piece later and can ensure a streamlined workflow for caregivers.
VDI Prerequisite Requirements
Assess, Design and Plan
A high-level review of the hospital's requirements will need to be undertaken to understand the number of endpoints, users, applications, and printers per department. Within the assessment and design stage, it is necessary to address the volume of users and endpoints required — these numbers are also needed for procurement as they will dictate hardware and licensing counts.
Key requirements to identify:
- Applications needed
- Number of endpoints
- Number of users
- Printing solutions
Beyond the high-level requirements gathering, a detailed review of clinical workflows should be created to document the risks and challenges associated with deploying VDI. Key areas to assess include:
- Applications — Some applications are not compatible with VDI and will need special attention. Common challenging applications include vital sign monitoring, pre-natal monitoring, and image viewing.
- Printing workflows — Look at all application printing workflows, not just Windows printing. Many healthcare applications use complicated printing workflows based on triggered events or have printers mapped within the application.
- Staffing and float teams — Understanding teams that float through departments allows better sequencing and grouping of go-lives.
- Generic logins — Significant change to workflow will be required in departments that use generic logins for desktop access.
- Special use case workstations — VDI is not an all-or-nothing initiative; certain workstations for imaging or monitoring may be excluded.
- Local file storage — All files stored locally on PCs needed by staff must be migrated to a network drive or SharePoint prior to rollout.
After the hardware analysis, a VDI environment design needs to be created covering: host utilization (CPU, RAM), storage, images/application delivery, policies (AD, VMware/Citrix), remote connectivity, Wi-Fi coverage/IDF capacity, endpoint requirements, failover and fault tolerance, and endpoint naming conventions.
T2 Tech recommends each department be allocated a minimum of one to two weeks for deployment depending on the current state environment.
Build
The VDI build phase requires coordinating with engineers to assign tasks, understand availability, and document each build component. Key build prerequisites include:
- Rack, cable, and power hardware
- Build VDI server environment (infrastructure servers, database servers, external gateways)
- Build VDI parent image
- Build application delivery tools (XenApp, AppVolumes, etc.)
- Build VDI policies and end-user profiles
- Build SSO system and application profiles
- Build printing system and queues
- Build endpoint management system and configuration
The parent image build requires careful planning as it represents VDI's core end-user experience. A core objective for any VDI image is to keep it as lightweight as possible while ensuring all necessary functionality is met. It is recommended to minimize agents and drivers and utilize application delivery tools like Citrix's XenApp and VMware's AppVolumes instead of locally installing applications on the image.
Once the parent image is created, T2 Tech recommends using a floating desktop configuration instead of persistent desktops. This provides each user a new, optimized desktop on a scheduled basis, refreshing nightly so staff start their shift with a clean environment.
End-to-End Testing
The testing process needs to be structured to vet each aspect of the newly implemented system. Testing functionality and redundancy of the new virtualized environment will identify possible issues within the initial architecture. High volumes of user traffic during shift changes can create a bottleneck and must be tested to validate if the platform can withstand the necessary load.
- Create and document test scenarios for all end-user workflows, including individual testing of each application, peripheral, and printing workflow
- Third-party tools may be needed to facilitate load testing and boot storm testing
- Create an endpoint lab environment and test scenarios
- Document login and roaming times
- Troubleshoot and implement identified resolutions
Five-Step VDI Deployment Strategy
The following five steps are led by an agile approach which incorporates the principles of vertical slicing — fully completing each departmental rollout before moving to the next.
Step 1: Communicate and Train
A clear communication plan needs to be created to help alleviate any stress and anxiety users may have regarding the technological transition. T2 Tech recommends:
- House-wide communication in newsletters
- An intranet site with project information, benefits, ROIs, training material, and high-level timeline
- Department-level one-on-one meetings to walk leadership through the rollout plan
- Attending morning standup meetings with each department prior to rollout
- Signage in departments identifying whether a department is live with VDI or planned to go live the next week
End-user training should include a training curriculum and script, comprehensive training handout, and a training video or tutorial. Training can take up to ten minutes per user and includes SSO badge enrollment, what will look different, workflow changes, common questions, and secure printing workflow.
Step 2: Document Department Specific Needs
Before deploying to each department, a detailed current state assessment and future state plan needs to be documented. This goes into considerably more detail than the initial assessment. Document per department:
- Current endpoint names and endpoint type (desktop, laptop, WOW, PACS viewing machine, etc.)
- Location of each endpoint
- Future endpoint name with naming convention adjusted to refer to a VDI endpoint
- Applications loaded on each workstation
- Peripherals used on each workstation
- Mapped printers for each workstation
- Network ports used by each workstation
Step 3: Build Department Specific Requirements
The items identified during Step 2 will serve as a blueprint for each department. Depending on the components identified, the following will be completed:
- Provision user accounts and deploy SSO profiles
- Configure and map department specific applications
- Build department specific printers or secure printing devices (if applicable)
- Build, name, and configure department specific endpoints
Step 4: Deploy Endpoints and Go-Live
A small to medium sized department should be chosen to pilot the new solution. The pilot program should run for a minimum of two weeks, issue free, before moving onto the next department. Key efforts include:
- Select a good use-case department
- Enrollment and training of users
- One week of dedicated support following go-live
- Monitor for another week
- Identify issues and resolve
- Observe and improve workflow
- Two weeks issue free (recommended)
- Reflect and validate deployment strategy
It is recommended that each department deploy their required set of endpoints before the start of a new shift. Once deployed, end users should have dedicated support for the following 48 hours.
Step 5: Remove Old Workflow Equipment
After the virtualized environment is live and issue free for at least two weeks, it can be determined which desktops can be removed from the network.
Reflect and Repeat
Lastly, take the time to reflect on successes and failures. Document ways to improve for the next department deployment and repeat the process for each remaining department. This iterative approach — completing each departmental rollout fully before moving to the next — is what ensures a consistent, high-quality clinical experience across the entire organization.
Authors
Kevin Torf — Managing Partner, T2 Tech Group. 30+ year career as an information systems executive specializing in large-scale IT project design, procurement, and implementation including data centers, server farms, storage and backup systems, security, and VoIP systems.
Leigh Sleeman — Partner, T2 Tech Group. 17+ years' experience as a program manager in healthcare IT, having championed large VDI, storage, and network initiatives for Kootenai Health and a datacenter migration for Sharp HealthCare. Former IT PMO Manager at UCLA Health Systems.
Kyle Torf — Program manager and healthcare technology expert with experience in enterprise-scale IT infrastructure transformations, VDI/SSO rollouts, big data strategies, population health solutions, and interactive patient care technologies.
Ready to explore a partnership?
Let's Connect