Case Study
Tempus Drives Remote Office Reliability and Resilience through VDI Automation
Trigent automated 450+ VDI instances using Terraform, enhancing operational efficiency, site reliability, and security
About the Client
The client, Tempus, is a leading AI-driven health tech company pioneering innovative solutions for healthcare and research. It focuses on leveraging technology to enhance patient outcomes. Over the years, it has become a global leader, driving advancements in precision medicine and personalized healthcare. With a commitment to excellence, it continues pushing boundaries, transforming disease understanding, diagnosis, and treatment worldwide.
Business Challenge
Tempus has several HealthTech customers globally, who are provided secure access to its platform and data for clinical research, life science projects, and drug development. To ensure data security and integrity during remote access, its customers need to use a Virtual Desktop agent image on their existing workstations, connect to a secure Virtual Private Cloud (VPC), launch the Virtual Desktop Interface (VDI), and log in using the secure credentials provided by Tempus. Based on the user profile, apps and services authorized for that user are available. Tempus’ images are generated daily to ensure data integrity and security.
Tempus allows two of its healthtech customers to access its platform via 450+ Virtual Desktop environments using AWS Cloud WorkSpace configurations. WorkSpaces allows customers to securely access its vast library of clinical and molecular data, apps, and other vital resources for their healthcare research and product development.
Tempus wanted an efficient way to manage the remote infrastructure, regulate access, and monitor performance, which enhanced site reliability and service availability. Therefore, it wanted to automate the management of the VDI infrastructure completely.
Tempus chose Trigent as its tech partner based on our extensive experience in infrastructure services.
Trigent Solution
Trigent took a comprehensive approach to addressing Tempus’ challenges. As part of our discovery exercise, we understood they leveraged a multi-account AWS architecture for a secure, cost-effective solution. AWS WorkSpaces resided in a dedicated account for centralized management. However, resources critical to WorkSpaces were spread across various other accounts. To seamlessly connect these disparate elements, AWS Transit Gateways were deployed. These transit gateways acted as central hubs, enabling secure communication between the WorkSpaces account and the resource accounts. This design facilitated granular access control, improved security posture, and optimized cost allocation by isolating workloads.
We established a complete staging environment tailored for local developer use. By configuring a staging VPC, we ensured seamless accessibility for VDI instances to the internet and private resources of Tempus, facilitating efficient development and testing processes for local developers.
AWS WorkSpaces Automation
Leveraging Terraform – an Infrastructure-as-a-Code software, we codified the entire AWS environment to automate the setup and configuration of various AWS resources. This encompassed VPC configuration, Lambda functions for workspace management, and the import of various resources, such as IP Access Controls and AD Connectors, across diverse regions. This approach ensured consistency, speed, and scalability in infrastructure deployment, enabling efficient management across multiple regions. Daily rebuilds updated the Skylight agent to ensure the security and performance of the VDI environment.
The Trigent Ninja team created a Bash Script that automated the import of over 450+ VDI configurations into Terraform. This ensured efficient management without manual intervention in addition to uniformity and scalability in infrastructure deployment, translating into overall site reliability and availability.
This creative approach resulted in an 80% reduction in time and effort, allowing the team to complete the process in 10 days with fewer resources. Tempus appreciated it, and it now serves as a best practice reference for site reliability operations for all their customers.
Furthermore, we conducted a Proof of Concept (POC) for Workspaces Web, enabling its global users to access WorkSpaces via a secure web interface instead of the traditional client. This innovative solution aimed to enhance user experience and streamline access to Tempus’ websites for its offshore clients.
As the project advanced, we fine-tuned and refined our approach by optimizing network performance and providing comprehensive end-user support. We integrated essential tools like AWS Console, Jira, and GitHub to facilitate collaborative efforts and streamline workflow management. Additionally, utilizing Tempus’ in-house tools like TAM and okta-personal-token, we strengthened the operational framework, offering indispensable support in managing and monitoring the VDI environment.
CloudWatch Alarm and Monitoring System
As a proactive measure, we went beyond the initial requirements to implement proactive monitoring and automated notifications that added to the Site Reliability. We set up a smart CloudWatch alarm and monitoring system across all VDI instances. It continuously monitors key metrics like Insession Latency, enabling early detection and resolution of potential issues and enhancing overall system reliability. In addition, we ensured everyone stayed in the loop by setting up alerts through emails and Slack channels. With PagerDuty – an incident management tool – in place, critical issues get flagged to our phones for a speedy response.
We also developed comprehensive guidance documents for the entire infrastructure setup, creating a shared playbook for future reference.
Client Benefits
Trigent’s collaboration yielded a refined technology stack, optimizing VDI usage and enhancing user experience.
- Centralized Infrastructure Management: By creating Terraform code for all AWS WorkSpaces resources, management was consolidated within the organization's main GitHub repository. Automation enabled Tempus to onboard new resources or make configuration changes easily and quickly.
- Proactive Issue Detection: Establishing alarm systems across the entire infrastructure promptly alerted relevant stakeholders for effective resolution for enhanced site reliability.
- Strengthened Security: Crafting new images and bundles with the latest security patches for Windows servers bolstered data protection measures.
- Swift Problem Solving: We minimized service disruptions and ensured uninterrupted operations by enabling faster identification and resolution of a global downtime issue in one of its global locations within three days.
- Saved Time and Effort with Terraform: The novel method to import 450+ WorkSpaces into Script Code saved 80% of the time and effort and is now the best practice for all VDI deployments.
- Reduced Support Overhead: CloudWatch alarms significantly decreased end-user support emails, enhancing Tempus’ operational efficiency.
- Streamlined Access for Offshore Users: With Workspaces Web, Tempus improved user experience and simplified access to its internal websites for offshore teams.
- Enhanced Support Efficiency: With Trigent handling the complete support and troubleshooting responsibilities, Tempus could deliver seamless real-time operations.
Technology Stack
