How INDstocks Built A Trading Engine Designed For Real-time Markets?

When you are building a trading platform in India, you are not just competing on features but on milliseconds. Every morning at 9.15, markets open and within a matter of seconds, the entire days’ pent up trading activity floods in at once. However, this chaos is not confined to market opening. 

It repeats throughout the day, whenever breaking news hits, policy announcements drop, or corporate actions move prices.  When we built INDstocks, we architected for exactly this kind of chaos, not just average traffic. Our goal was clear: give traders the speed and reliability they need to act on market opportunities without the platform getting in their way.

Amazon Web Services (AWS) recently published a case study on how we built the INDstocks trading engine that delivers trades and processes orders in less than 20 milliseconds and delivers price ticks from the exchange to traders’ Apps in under 50 milliseconds using Aurora PostgreSQL, AWS Multicast Groups and Global Accelerator. This is the story behind those numbers, the technical decisions and what we learned building a trading engine designed for real-time markets.

The Real Challenge: What Trading Platforms Must Solve

Any trading platform needs to have these three non-negotiables:

Speed: We knew the success of our platform directly depended on how fast the data was updated on the app. If price moves and a trader does not see it even for a second, they are trading blind, which could lead to bad decisions. To solve this, we needed real-time price ticks from the exchange to the traders' App.

Scale: The stock market is known for its volatile nature. This means the market does not rally or tank gradually; this change happens instantaneously. Budget announcements, policy changes, and earnings reports are only some of the events that lead to this spike. We needed an engine that could handle sudden traffic spikes without degrading performance and without over-provisioning Infrastructure.

Security: Security is not optional. It goes without saying that financial platforms handle some of the most sensitive data. We knew we needed a trading engine that is rock-solid against any external threats or unauthorised access.

The Foundation: Cloud-Native Architecture on AWS

Our structure was laid. We now need to build the foundation that offers Indian traders and investors a simple and secure way to trade and invest in the share market. 

While INDmoney relied on AWS for its broader backend, for the INDstocks trading engine, we needed a more deliberate service selection. To do this, we built the trading engine on a cloud-native architecture using Amazon Web Services (AWS).

To focus on speed, security and scale, INDstocks uses the following services by AWS:

1. Amazon Aurora PostgreSQL

Amazon Aurora PostgreSQL became the backbone of our order processing system. It scales through multiple readers when markets open. This meant we could handle peak loads when markets opens and tens of thousands of trades hit simultaneously.

2. AWS Multicast Groups

When a price update comes from NSE or BSE, multiple systems need that data, from order management systems to the WebSocket servers.  Every hop is an additional latency overhead. Our Engineering included AWS Multicast Groups, which ensure all our services receive the ticks at the same time, without any extra hops, bringing a significant latency advantage to users.

3. AWS Network Load Balancer

The next challenge was that real-time price streaming required persistent WebSocket connections. This means every traders’ App stays connected to receive live price ticks and their respective trade updates. Instead of putting our Order and Price Tick websocket servers behind a regular Application Load Balancer, we chose to use AWS Network Load Balancer. NLB helps us scale seamlessly and again gives a latency advantage to the users because it operates at the Transport Layer of the OSI model, handling millions of TCP / UDP requests at ultra-low latency. AWS Auto Scaling takes this further by monitoring traffic in real-time and spinning up additional servers during surges.

4. AWS Global Accelerator

India’s internet infrastructure varies by location. Traders in remote parts of the country are often faced with latency spikes due to inconsistent internet infrastructure. To solve this, Global Accelerator routes traffic over the AWS Global Network to the closest, best-performing endpoint, reducing latency significantly.

Secure Infrastructure: Building Trust In Every Layer

Building a secure infrastructure has been our topmost priority from the start. It is not just a layer we added, but it is embedded in every architectural decision. This led us to integrate the highest grade security services from AWS, including:

Amazon GuardDuty

This helps us with continuous threat detection across our AWS environment. Think of it as a guard analysing all activity across your AWS infrastructure, that alerts you on any suspicious activity instead of waiting for a breach to happen.

AWS Secret Manager

This service handles all credential storage from API Keys to database passwords and third-party tokens. 

Amazon Inspector

This service runs automated vulnerability assessments on our infrastructure. It scans for network exposure, software vulnerabilities and deviation from best practices.

What Does This Mean For Traders?

We set out to build infrastructure that traders would not have to think about. The goal behind INDstocks was to develop technology that operates quietly in the background, allowing traders to focus on their decisions. AWS gave us the infrastructure, and our teams delivered on it with a constant eye on metrics-driven engineering - making sure every data packet and associated latency is measured and decisions are supported by data and not by assumptions.

Disclaimer:
This blog is for information/educational purposes only. Investments in the securities market are subject to market risks. Read all the related documents carefully before investing.  INDstocks Private Limited (formerly known as INDmoney Private Limited) 616, Level 6, Suncity Success Tower, Sector 65, Gurugram, 122005, SEBI Stock Broking Registration No: INZ000305337, Trading and Clearing Member of NSE (90267, M70042) and BSE, BSE StarMF (6779),  Trading Member of MCX (57600), AMFI Registration No: ARN-254564, SEBI Depository Participant Reg. No. IN-DP-690-2022, Depository Participant ID: CDSL 12095500, Research Analyst Registration No. INH000018948 BSE RA Enlistment No. 6428.