Flipkart - India's biggest E-commerce Platform
Deepanshu Rastogi
- Scaled the MySQL database cluster [8 shards] and optimized queries to support over 400k QPM (with a peak read rate of up to 170k QPS and peak write rate of 8k QPS) per shard, running load tests on the system to optimize and identify bottlenecks for India's biggest e-commerce sale event, "Big Billion Days".
- Led requirement-gathering sessions with Technical Architects and Product Managers to collect information related to peak and average orders per hour for major sale events.
- Conducted brainstorming and requirement-gathering sessions with external teams to identify API SLA changes and API call growth rates by converting orders per hour to API calls and database queries per second across the microservices we own. This approach helped in formulating a plan to identify bottlenecks, conduct load testing, and scale the system according to business needs.
- Developed tools to move data from the transactional store to the cold store, which, after successful testing, were exposed as an API interface to transfer data between the hot store and cold store as per business requirements.
- Participated in the grocery business scaling charter and helped scale the grocery business in Tier 2 and Tier 3 cities by utilizing existing non-large supply chains.
Like this project
Posted Apr 21, 2023
Worked on the scalability of the backend application, and improved performance for an expected exponential increase in traffic due to sale events.
