Alex Kimber, a principal technologist at AWS, discusses the use of high-performance computing in financial services for risk modeling. Topics include leveraging HPC for measuring credit, market, and actuarial risks, AWS services enabling HPC workloads, HTC Grid development, Lambda capabilities, optimizing HPC with AWS storage solutions, and navigating AWS services for financial services system re-architecture.
Financial firms leverage HPC for risk modeling to measure/manage credit, market, and actuarial risks.
AWS provides HPC architectures like HTC Grid for massive scalability and performance optimization in financial services.
Deep dives
Use of High-Performance Computing in the Financial Services Sector
Financial services companies are utilizing high-performance computing (HPC) for risk modeling, including credit risk, actuarial risk, and market risk. By effectively modeling risk, these companies can better price their products, manage risks, and comply with regulatory requirements. The complexity of risk models requires extensive scenario calculations, such as Monte Carlo simulations, which benefit significantly from large-scale HPC systems.
Contrasting Workload Types in HPC: Financial Services vs. Weather Modelling
HPC workloads in financial services often involve loosely coupled tasks that run independently, leveraging parallelization for increased efficiency and horizontal scalability. This differs from tightly coupled tasks like weather modeling, where task outputs influence each other and necessitate rapid communication between tasks. AWS provides services tailored to both workload types, optimizing performance for financial services through parallelization and scalability.
Challenges and Optimizations in Financial Services HPC
In financial services HPC, instance startup time plays a crucial role in overall workload performance. Addressing factors like security agents deployment, data distribution, and proprietary analytics code optimization can significantly reduce initialization times. By optimizing instance storage choices and leveraging AWS services like S3, FSX for Lustre, and EBS, customers can enhance speed and efficiency in large-scale financial HPC calculations.
Join Simon as he discusses the use of high-performance computing (HPC) in the financial services industry, particularly for risk modeling and regulatory reporting. Alex Kimber, a principal technologist at AWS, explains how financial firms leverage HPC to run millions of scenarios to measure and manage risks like credit, market, and actuarial risks. The workloads are "embarrassingly parallel," allowing for massive scalability in the cloud. Kimber also discusses AWS services and architectures that enable these HPC workloads, including the HTC Grid, a blueprint for a simple, high-throughput scheduler on AWS.
Get the Snipd podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share & Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered podcast player
Listen to all your favourite podcasts with AI-powered features
Discover highlights
Listen to the best highlights from the podcasts you love and dive into the full episode