The middleware initiative at HDFC Bank


Middleware played a major role in advancing HDFC Bank’s core banking services with streamlined analytics capabilities and real-time insights. Learn more here.

Business context

As one of the largest banks in India, HDFC Bank currently has around 76 million customers, of which 4 million have enabled net banking and mobile banking. The concurrent usage rate of the digital platform is around 50,000 transactions per second.  Understandably, the Bank’s customers have high expectations and want to conduct and view their banking operations as quickly as possible. When a customer logs in to the platform, the system fetches the customer’s details from the core banking system and displays the relevant information required by the platform.

Solution deployed

In between the core banking and net banking systems lies the Oracle Banking Platform– the middleware in this equation. When it comes to middleware like OBP, the aim is to build the processed data as a result-set cache in Coherence. This is done in close proximity to the OBP for a more streamlined process, so that each request can be addressed in the middleware layer, without having to go back to core banking databases to fetch the data. When establishing the viability of this type of middleware, the following points were always at the forefront of Oracle and HDFC Bank’s minds:

  • The expected response time should be under 25 milliseconds
  • Coherence should be able to handle the data of over 4 million customers
  • Logic should be implemented to expire the cache on any changes in core banking
  • Coherence should be designed to ensure 100% uptime


Today, Coherence is used effectively in real time to deploy solutions built on Oracle Stream Analytics. So, its implementation in other areas shows great promise. In fact, Coherence also could be applied to many other product processors where the data is static, such as payment applications like ‘eSource 2.0’. The team is currently exploring all the possibilities.

Share this Page