Pharmaceutical

Faster Drug Modelling for Global Pharmaceutical Leader through AWS HPC

By migrating the PK/PD simulation infrastructure to an AWS-native HPC environment, Invisibl Cloud enabled a transformative uplift in drug discovery workflows—achieving 5x faster simulations and 20% lower compute costs
Faster Simulation Run Times
Job Run Cost Reduction
of Parallel Simulations

The Challenge

A global pharmaceutical leader faced critical bottlenecks in their drug discovery pipeline.

mingcute_time-line

Infrastructure Dependency Delays

Drug discovery process & clinical product designs were delayed due to dependency on infrastructure requirement to run simulations in traditional VM environments.

ion_people-orange

Limited Researcher Productivity

Researcher productivity was constrained by infrastructure that lacked support for running parallel simulations and analysis workflows.

gg_database

Lack of Scalable Platform

Absence of a scalable platform to serve multiple research users who needed to run multiple simulations in parallel simultaneously.

Our Solution

End-to-end deployment of a production-ready AWS HPC environment optimized for running PK/PD workloads using NONMEM.

HPC Foundation Setup

Implemented an AWS ParallelCluster based HPC environment directly within the customer’s AWS account for full data control and enterprise security.

HPC Cluster Configuration

Configured SLURM HPC cluster with optimized queue and QOS setup adhering to AWS Well Architected practices.

Application Integration

Installed and validated NONMEM, PsN, and FinchStudio on the AWS HPC stack, ensuring seamless execution of bootstrap, estimation, and simulation workloads.

User Enablement

Onboarded platform admins and HPC users with comprehensive self- service capabilities including job submission, monitoring, and governance.

HPC Capabilities

Transformational Results

The AWS-native HPC implementation delivered exceptional improvements across all performance metrics.

5x

Faster Simulations
(120h → 24h)

20%_b

Lower Compute
Cost

100s

Seamless Scalability to 100s of HPCCompute Nodes

Zero

Runtime
Failures

Pharmacometric Workflow Capabilities

Job Submission & Monitoring

Submit NONMEM jobs, run parallel simulations, and monitor logs and performance in real time via the HPC workflow environment.

Budget Management

Track compute spend, configure usage alerts, and optimize cost efficiency across research studies.

Reporting & Analytics

Generate detailed simulation reports and resource utilization insights for scientific and operational teams.

Governance & Compliance

Implement workload governance, enforce HPC usage policies, and maintain compliance across the cluster.

Fast Facts

Technologies

NONMEM & PsN

AWS EC2

Amazon S3

SLURM Scheduler

FSx for Lustre

RDS

FinchStudio Integration

AWS Parallel Cluster

Ready to Accelerate Your Research?

Discover how Tachyon can transform your drug discovery pipelines and unlock new levels of research productivity.

Learn how Apollo Tyres leveraged Tachyon to scale engineering simulations & drive innovation​