WORK
Seamless Onboarding of Algo-Trading Customers
Business Problem
The Client wanted to enhance legacy P&L generation platform that was plagued with multiple issues:
Onboarding of high-volume Algo-trading customers: High-volume transactions were processed through traditional RDBMS based DB system impacting onboarding.
- Scalability: Existing architecture relied on a legacy database for all data extraction use cases. Application scaling was a challenge, and this put the database under heavy stress while processing the transactions.
- Real-time reporting: Lack of a robust reporting solution limited the visibility into various processing parameters.
The Solutions
Scalable Performance
Solution built on a “distributed cache system” to cater to extremely high volumes (>500K rows) of transactions.
Real-time Analytics
Built a new In-memory data grid solution that aggregates & transforms the data in-memory and provides paged results and in real-time.
Configurable Snapshots
Customized snapshot data for a configured period from the current business day.
Adaptable Pricing Solution
A pricing solution was built using micro services and NIO architecture to cater to a huge volume of products and the ability to configure new price sources.
Efficient Data Integration
Solution facilitated the data push to the downstream systems.
Value Delivered
Reduced latency of High-volume reporting to less than 5 seconds..
2M
Requests per day handled
500K
Transactions per minute processed
Significant improvement of UI/UX experience of the screens because of integration with the new platform.
