Hoonify Transforms Radioisotope Simulations with TurbOS® Software

Background

Modern nuclear simulation research can be accelerated with high performance computers which employ parallel computing powers to run trusted programs including the Monte Carlo N-Particle (MCNP) application. To enable this critical research, Hoonify was endorsed by Oak Ridge National Laboratory’s Radiation Safety Information Computational Center (RSICC) to host secure access to MCNP for select, authorized users. RSICC’s strict security compliance and data-handling processes necessitate greater cyber security protections, something which Hoonify provides through our TurbOS platform. Other cloud solutions typically cannot meet these security and compliance requirements, which is why Hoonify is one of the only private server operators endorsed by ORNL’s RSICC.

Northstar Medical Radioisotopes is a global innovator in the development, production, and commercialization of radiopharmaceuticals used for therapeutic and medical imaging applications. Their mission has positioned them as an emerging leader at the forefront of U.S. medical radioisotope research, heavily relying on computational modeling, such as MCNP simulations, to optimize their development services.

The Challenge

NorthStar needed faster, repeatable MCNP results to keep product and regulatory timelines on track. Multi-day simulation times created a bottleneck for these results, which ultimately limited the number of patients helped. Older, on premises computing resources limited the space and power available for newer, faster solutions, and IT’s limited HPC expertise was preventing the team from fully utilizing the parallel computing capabilities of MCNP. Public cloud services introduced added cost and compliance concerns, creating additional business risk and unpredictability. 

• No cost-predictable path to scale - On-prem expansion required new power and cooling; public cloud introduced unpredictable pricing for ongoing simulations.

• Slow simulation results. Unoptimized computing resources created inefficiencies with MCNP simulations, often running for days and delaying research insights.

• Throughput limited by capacity and tuning. Aging computing clusters and workstations, combined with IT’s lack of high-performance optimizations (compilers/libs, MPI/OpenMP, I/O) created frequent bottlenecks.

• Expertise and compliance gap. The team lacked specialists with HPC experience to stand up a compliant, high-performance, optimized environment for MCNP; manual admin increased operational complexity—pointing to a managed solution.

Business impact: Without access to same-day answers with current staffing and compute resources, NorthStar experienced delayed results, fewer design points per week, and mounting costs—negatively effecting product readiness and customer commitments.

The Solution

Hoonify delivered TurbOS via a private cloud solution optimized for MCNP—combining parallel performance optimizations, managed operations, and predictable capacity—so MCNP simulations delivered same-day results without additional staffing, overhead or computing resources.

MCNP Simulation Input

NorthStar uses existing MCNP models and inputs (geometry, materials, physics cards, particle histories).

• Secure upload and authorized application access
• No changes to user workflow

TurbOS™ Parallel Processing & Scheduling

TurbOS maps MCNP’s parallel model to our TurbOS platform optimized for immediate efficiency.

• Resource allocations, job scheduling, workflow management
• Optimized compilers, libraries, and memory placement with repeatable runbooks
• Simple dashboard for interactive operations – without requiring IT admins

Private Cloud

Dedicated capacity sized for MCNP throughput—without adding on-prem racks.

• High-density compute nodes with fast scratch/parallel storage
• Predictable cost model with dynamic scalability when needed
• Isolated environment with enhanced security controls and auditability

Simulation Output

Same models, but dramatically faster simulations and more predictable results.

• 96× faster results (4 days → 1 hours) with on-demand access

• 50% lower cost vs. standard, public cloud alternatives

• On-time milestones with zero additional staffing or compute resources

The Implementation

NorthStar moved from older, on-premises computing resources to an optimized private cloud solution in one day. The steps below mirror the “Rapid Implementation Timeline.”

1. Scope & Planning - Confirmed desired results timelines, simulation workloads, success criteria, and compliance requirements. Finalized the plan for transitioning to a private cloud solution, partnership roles, and day-by-day schedule.
   
2. Access Setup - Provisioned secure access, configured firewall security and data paths, and enabled cyber security auditing requirements so researchers could submit MCNP jobs securely on day one.
   
   
3. TurbOS Deployment - Installed and configured TurbOS on private cloud resources, including SLURM for efficient use of resources, MCNP modules and containers, optimized storage, and automation tools.
   
   
4. MCNP License Integration - Validated authorized software licenses, enforced license-aware access controls, and documented the compliant workflow for ongoing operations.
   
   
5. Training & Optimization (Go-Live) - Delivered hands-on training and job templates for smooth operations after validating increased simulation performance and going live.

The Results

NorthStar saw a nearly 100x improvement in simulation results, allowing NorthStar engineers to increase their total simulations from 60% to 80% and reduce IT involvement by 20%. These improvements significantly accelerated NorthStar’s research results, helping them improve their competitive edge in this growing industry.

"Time is our biggest enemy. Having an optimized system that provides accurate results and a 100 percent improvement in simulation performance is critical. Analysis is the bottleneck in our pipeline, and this improved HPC performance significantly enhances our workflow efficiency, helping NorthStar innovate faster and maintain our competitive edge."
– Josh Peterson-Droogh, Nuclear Engineering & Physics Manager at NorthStar Medical Radioisotopes

Operational Simplicity & Productivity

Leveraging Hoonify TurbOS enabled NorthStar to significantly reduce the time for them to get critical results, increased its ability to provide accurate computational results to benefit patients and research activities, provided an easy-to-deploy and easy-to-use HPC capability that they never had before, and reduce its internal IT overhead. This allowed the research teams to remain focused entirely on scientific discovery and innovation, and it enabled IT to focus on greater enterprise projects and continual service improvements.

Introducing TurbOS Dash

TurbOS Dash provides a simplified interface for engineers and analysts to perform their complex simulations and research, including:

1. Start Smart: Simple landing page displays your key metrics
   

2. Better Visibility: Usage and Cost in One Report
   
   
   
3. Better Visibility: A Single Page for Node Management