Zerodha cuts backend processing time from hours to minutes using AWS Batch

Zerodha, India’s largest retail stockbroker by active client base, has significantly enhanced its backend processing efficiency through the adoption of AWS Batch.

The service enabled the company to reduce the execution time for critical calculations, such as daily profit and loss (PnL) computations, from several hours down to just minutes. This was crucial for Zerodha to meet its business objective of providing accurate daily PnL calculations to its customers amidst growing data volumes and evolving market demands.

Traditionally, backend processes for stockbrokers, including settlements and invoice generation, are CPU-intensive, requiring the processing of vast datasets that can span several years. Zerodha faced escalating challenges in managing these processes due to its rapid expansion in customer base and transaction volume.

The company moved away from previous third-party commercial software solutions that offered limited optimization capabilities. By leveraging AWS Batch, Zerodha was able to dynamically scale its computing resources to meet processing demands, leading to substantial improvements in both performance and turnaround time.