Case Study

Global Bank Replaces Vendor Platform with ORE + AADC

From $3M+ annual fees and overnight batch to real-time risk at 90% lower cost

The Challenge

A global investment bank faced escalating costs and limitations with their incumbent vendor platform for derivatives pricing and risk:

$3M+ annual license fees with regular price increases
Batch-only risk calculations overnight runs, no intraday visibility
Limited model control waiting months for vendor to implement regulatory changes
GPU infrastructure pressure vendor pushing expensive GPU migration for performance
Vendor lock-in proprietary formats, difficult to extend or customize

The bank needed real-time risk capabilities for trading desk decision-making, but the vendor's proposed GPU upgrade would cost millions more with continued dependency.

The Solution

The bank chose Open Risk Engine (ORE) as their open-source foundation, accelerated by MatLogica AADC:

ORE provided proven, transparent pricing models

Full source code access and complete control over valuation logic

AADC transformed ORE's performance

From batch-suitable to real-time capable with 600x acceleration

Unified platform for XVA and Live Risk

Same codebase serves both calculations - no duplication

The Results

Metric Before (Vendor) After (ORE + AADC)
Portfolio Pricing Minutes Sub-second (600x faster)
Risk Sensitivities Overnight batch Real-time (600x faster)
Annual Cost $3M+ licenses ~90% reduction
Model Changes Months (vendor dependency) Days (in-house control)
Hardware GPU upgrade proposed Standard CPUs only
Vendor Lock-in High None

Technical Highlights

600x acceleration on standard AVX512 CPUs
<1 Adjoint factor - Greeks faster than pricing alone
Auto Automatic AAD - no manual derivative coding required
Mix Bank's proprietary models integrated with ORE

Key Benefits Achieved

1

Real-Time Risk Enabled

  • Traders now see live portfolio risk throughout the day
  • What-if scenario analysis in seconds, not hours
  • Intraday limit monitoring and breach alerts
2

$3M+ Annual Savings

  • Eliminated vendor license fees
  • No GPU hardware investment required
  • Reduced cloud compute costs (same workload, fraction of resources)
3

Full Model Control

  • Regulatory changes (FRTB, SA-CCR) implemented in days
  • Custom model extensions without vendor negotiations
  • Complete transparency into pricing logic
4

Unified Platform

  • Single codebase for XVA and Live Risk
  • Consistent valuations across front office and risk
  • Simplified infrastructure and maintenance
5

No Vendor Lock-in

  • Open-source foundation (ORE)
  • Standard C++/Python skills (no CUDA specialists needed)
  • Freedom to extend, modify, or migrate
"We went from overnight batch risk to real-time trading floor analytics. The combination of ORE's transparent models and AADC's performance means we now have capabilities that exceed what our vendor offered - at a fraction of the cost and with full control."

- Head of Quantitative Development

Implementation Timeline

4 weeks

POC

Sample portfolio validation

4 months

Pilot

Single desk deployment

10 months

Production

Full rollout complete

Related Resources

ORE LiveRisk Technical Details

How to Turn ORE/QuantLib into a LiveRisk Service for FX Products: 1 Million Trades Priced in 0.4 Seconds with Complete Delta Risk

Learn More

How AADC Works

AADC extracts a computational graph (DAG) from existing analytics at runtime and compiles it to optimized machine code, achieving 100-1000× for Python and 6-100× for C++ with automatic derivatives.

Learn More

Deployment Strategies

AADC binary kernels run anywhere - from individual developer machines to enterprise data centers to cloud infrastructure. Choose the deployment strategy that matches your security requirements, performance needs, and existing infrastructure.

Learn More

Ready to Achieve Similar Results?

Let us show you how ORE + AADC can transform your derivatives infrastructure

Schedule Consultation View More Case Studies