Uber System Design | Ola System Design | System Design Interview Question - Grab, Lyft

Summary

This video script outlines the design of a cab booking system akin to Uber, focusing on functional requirements like cab visibility, trip estimation, and booking, and non-functional aspects like global accessibility, low latency, high availability, and consistency. It delves into the architecture, including services for user and driver management, location tracking, trip handling, and real-time communication via WebSockets. The script also discusses using data for analytics, driver and customer profiling, and enhancing services like ETA calculations.

Takeaways

• 🚖 The script outlines the design of a cab booking system similar to Uber, OLA, or Lyft, focusing on functional and non-functional requirements.
• 📍 A primary functional requirement is the 'See Cab' feature, allowing customers to view nearby cabs through the app.
• 🕒 The system should provide an estimate of the travel time and cost from point A to point B for customers planning their trips.
• 🔍 Users should be able to book a cab easily, with the system supporting a single type of cab initially, but designed to be scalable for additional types.
• 📊 The system must include robust location tracking for drivers to monitor their positions and movements in real-time.
• 🌐 Non-functional requirements emphasize global accessibility, low latency, high availability, and high consistency, acknowledging the CAP theorem's limitations.
• 🔢 The design targets scalability to accommodate millions of users and thousands of rides daily, as exemplified by Uber's statistics.
• 🔍 A critical problem addressed is efficiently matching customers with the nearest drivers using a segment-based approach.
• 🏙️ The city is conceptually divided into segments to manage and locate cabs and drivers more effectively within the system.
• 📈 The script introduces a 'Maps Service' responsible for segment management, calculating ETA, routes, and distances, abstracting complex map functionalities.
• 🔗 The architecture includes various services like User Service, Driver Service, Location Service, and Trip Service, each with specific responsibilities and databases.

Q & A

What are the primary functional requirements for a cab booking system like Uber?

The primary functional requirements include the ability for customers to see available cabs in their vicinity, estimate travel time and cost from point A to point B, book a cab, and track the driver's location.

What is the significance of location tracking in a cab booking system?

Location tracking is crucial for determining the driver's current position and for calculating the estimated time of arrival (ETA) and route from the customer's location to the destination.

Why is it important for a cab booking system to have global accessibility?

Global accessibility ensures that the system can serve users in different geographical locations by having servers in multiple regions, which reduces latency and improves user experience.

What are the non-functional requirements that a cab booking system should ideally have?

The system should have high availability, low latency, and high consistency. It should also be scalable to handle a large number of users and transactions, such as the millions of rides per day that platforms like Uber handle.

How does the concept of 'segments' help in finding the closest drivers to a customer?

Segments are rectangular areas into which a city is divided. By determining the segment in which a customer is located and querying the surrounding segments, the system can identify the closest drivers, considering both aerial and road distances.

What is the role of the Maps Service in a cab booking system?

The Maps Service is responsible for dividing the city into segments, calculating ETAs, determining routes, and managing the logic for segment division and merging based on traffic and driver density.

How does the User Service interact with other services in the system?

The User Service acts as a repository for user information and a proxy to other services. It fetches user profiles, trip histories, and other information by interacting with services like Trip Service and communicates with the User App.

What is the purpose of the Driver Service in the system?

The Driver Service manages driver information, handles driver-related APIs, and communicates with the Driver App. It also interacts with services like Payment Service and Trip Service to provide drivers with information about their trips and payments.

Why is a WebSocket connection used between the driver and the system?

A WebSocket connection allows for continuous communication between the driver and the system, enabling real-time location updates and instant notifications about trip assignments without the need to establish a new connection each time.

What is the role of the Trip Service in managing trip information?

The Trip Service is the source of truth for all trip information. It manages live trip data in a MySQL database and archives completed trips in a Cassandra database. It also provides APIs for accessing trip details by ID or for a specific user or driver.

How does the system handle the assignment of drivers to customers?

The system uses a combination of the Cab Request Service, Cab Finder, Location Service, and Map Service to determine the best driver for a trip. The Cab Finder considers the customer's location, queries surrounding segments for nearby drivers, and uses the Driver Priority Engine to select the most suitable driver.

What are the potential uses of the events collected in the Kafka cluster?

The events in the Kafka cluster can be used for various purposes, such as initiating payments to drivers through the Payment Service, creating heat maps to guide drivers to areas with high demand, and powering analytics and machine learning models for user and driver profiling, fraud detection, and enhancing ETA calculations.