Summary
BioTeam partnered with a healthcare-focused investment firm, whose mission is to advance healthcare innovation by funding innovative, new biotech organizations. These organizations required an automated, secure, scalable, and repeatable cloud-based HPC environment on AWS for their drug discovery pipelines. The client needed a compliant technology foundation that could be deployed at scale across multiple ventures. BioTeam established a cloud-based, HPC infrastructure using infrastructure-as-code, enabling the automatic deployment of modern, holistic HPC environments.
BioTeam’s solutions enabled the client to reduce setup time for modern HPC environments, ensure regulatory-grade security, and accelerate research productivity.
Challenge
The client needed a reproducible, cloud-based HPC environment that enables researchers to efficiently copy and deploy standard genomic and protein structure analyses at a large scale, with security and cost controls already integrated. Priorities included:
- Enable repeatable and automated deployments across organizations.
- Integrate complex scientific applications, such as Schrödinger, with HPC schedulers to achieve efficiency, researcher access, and license awareness, thereby reducing contention and idle time on the HPC resources.
- Transition from a Windows- and Azure-only environment to Linux-based Virtual Desktop Infrastructure (VDI) and AWS-native HPC solutions.
- Proactively meet strict security, compliance, and governance standards.
- Reduce overhead by streamlining identity management.
Approach
BioTeam implemented a comprehensive cloud and HPC strategy:
- Infrastructure-as-code enabled BioTeam and the client to test and automate the consistent provisioning of modern HPC environments, integrated with scientific applications and license servers.
- Application build scripting using supercomputer package managers, automated deployment, and optimized complex scientific applications, such as Schrödinger and Posit, within the HPC system, ensuring license-aware scheduling.
- Built-in security & compliance were designed from the ground up within the AWS multi-account structure, utilizing Landing Zone Accelerator, SCP guardrails, and Fortinet next-generation firewall integration. BioTeam delivered a secure VPC architecture, including inspection and screening layers, to ensure a secure environment.
- Identity & access group mapping, along with comprehensive identity architecture within AWS, enforced the “Principle of Least Privilege,” thereby increasing the security of the cloud environment.
- Improve developer productivity by writing CI/CD pipelines, automating custom Amazon Machine Images (AMIs), and building repository templates to streamline development and enhance reproducibility.
- Improved end-user experience with Linux-based virtual desktop interfaces (VDI) for scientific workload applications, training sessions for researchers and administrators, and troubleshooting.
Outcomes
- Reduced HPC deployment time from weeks to hours through automation.
- Provided a secure, compliant, and reproducible AWS computing environment foundation for rapid use by new biotech organizations.
- Improved productivity with integrated scientific applications and license-aware HPC scheduling.
- Transition to AWS/Linux-native environments provided a standard, automated analysis environment with preferred research applications.
- Heightened security posture with enterprise-grade guardrails and next-generation firewall integration.
- Empowered end users and administrators with training, streamlined access, and scientist-friendly CI/CD workflows.
