.png)
Modernizing Financial Compute: From Legacy Schedulers to Cloud-Native Scale
Your on-prem grid is a bottleneck. It’s time to break the capacity ceiling.
For decades, financial institutions have been constrained by fixed-capacity compute grids - optimizing every millisecond of a static cluster to run overnight risk batches (VaR, CVA) or intraday pricing.
Join the High-Performance Compute experts from AWS to dismantle the old way of thinking. Learn how to architect a cloud-native High-Throughput Compute (HTC) grid that scales infinitely, eliminates the "queue," and processes massive volumes of tasks with low latency using the open source HTC-Grid project.
In this session, we will present HTC-Grid, an open-source high-throughput compute (HTC) grid system built on top of tried and tested cloud native services.
HTC-Grid incorporates learnings from many FSI customers re-inventing their grid systems for the cloud. The project was initially developed as an open-source blueprint by AWS, but has been donated to FINOS in order to share the obtained knowledge, provide a reference architecture that can be used to build and adapt a modern high-throughput compute solution, and to incorporate contributions from customers and other CSPs alike.
Who Should Apply?
This session is designed for the architects and quants responsible for the bank's most critical compute engines:
- Grid Architects & HPC Engineers: Who manage legacy grids (Tibco DataSynapse, IBM Symphony/LSF) and are planning cloud migrations.
- Quantitative Developers: Who need to understand how to write code that scales natively on cloud infrastructure (Lambda/Fargate/Batch).
- Cloud Platform Architects: Who need to build the underlying secure infrastructure to support bursty, high-volume workloads.
What You Will Walk Away With:
- Architecture Blueprint: A reference design for building a serverless, event-driven grid on AWS.
- Performance Patterns: How to achieve low-latency task scheduling at millions of tasks per hour.
- Migration Strategy: Practical advice on moving heavy quantitative workloads from on-prem monoliths to cloud-native micro-tasks.
Prerequisites
- Laptop Required: You will be deploying architecture.
- AWS Knowledge: Basic familiarity with AWS services (Lambda, SQS, DynamoDB) is highly recommended.
- Coding: Familiarity with Python or Go is a plus.
Logistics & Application
- Date: Monday, April 13, 2026
- Time: 1:00 PM – 5:00 PM ET
- Location: Toronto Financial District (Venue details provided upon acceptance)
- Cost: Free for approved applicants (Standard Value: $500)
Note: Due to the hands-on nature of this session, seating is strictly limited to 40 participants. Priority will be given to individuals from FINOS Member financial institutions and regulated industries.
