Requirements

Functional Requirements:

1. User can search the availability of the seats.
2. User Can book the ticket.
3. User Can get the notification on ticket booking and also get the reminder notification for the booking

Non-Functional Requirements:

1. Consistency: System must be highly consistent while booking a ticket and there should not be duplicate booking for a same seat. There can be the eventual consistency on the use notifications and getting a user booking details.
2. Availability: System must be available all the time.
3. Scalability: Application must be scalable in case of sudden spike on the popular movies ticket booking.
4. User should be able to search the events/movies with low latency

Capacity Estimation

Lets assume there are 100 million application user and 10% will be the daily active users. so there are 10 million active users out of which 10% actually book the ticket and other are read only users. So there will be 1 million user book the ticket daily and each user may book the 2-3 time in a day so there will be max of 3 million booking or writes in the system and if assume each user can search for a 4-5 movies and check for pricing and availability of the seat so there will be approx 50 million reads in the system if we consider 10 million active users daily.

API Design

1. get the latest movie list
   1. get the top 10 list of latest movies on the users dashboard
   2. HTTP GET /get-latest-movies
   3. As a response user will get the Movie details, Movie poster/image, ratings etc.
2. Search the movie by title
   1. User can search the movies using different criteria such as movie title, rating, location etc.
   2. HTTP GET /get-movie-detail
   3. Request: {movieTitle:"", date:"",location:""}
3. Get the seat availability for a specific movie
   1. Get the seat availability of a specific movie for a specific date and the location/cinema hall
   2. HTTP GET /get-seat-availability/{movieId}/{date}/{locationId}
4. Book the ticket.
   1. User can book the ticket by selecting the available seat for a specific movie for a specific date and the location/cinema hall
   2. HTTP POST /book-my-seat
   3. Request Body {movieTitle:"", date-time:"",location:"", seat:[{seatId:""}], user:""}
5. View the booking details.
   1. User can get the booking details.
   2. HTTP GET /my-booking-history/{userid}
6. cancel the booking
   1. User can cancel the booking. currently Inplace of modifying a booking user can cancel the existing booking and book a new ticket.
   2. HTTP PATCH /cancel-my-booking/{booking-id}

High-Level Design

There will be different services servers the system functionalities

1. search-service
   1. This service is responsible to server the movie details, availability details, and all filter criteria.
   2. It uses a distributed cache to get the latest movie details and the seat availability. on the cache miss it lookup into the databsae.
2. booking-service
   1. Booking service has a functionality to book the user ticket.
   2. It first checks the selected seat still available in the cache and if yes then it make a entry inside cache till the time user booking gets completed or max for 5 minutes. So here first seat availability will be verified and then reserve the seat for a user using a cache.
   3. For a payment it internally connects with the payment service from where the payment will be done and the cache seat availability cache will be evicted and the entry from the cache for reservation will be removed. it sends and booking event for a notification service from there further details will be share to user.
3. notification-service
   1. Notification of booking completed will be received to the service this service get the further details about the reservation from the database such as movie details, date time, location etc and prepare and send a notification to the user.
   2. It also has a schedulers which trigger after every 30 minutes and sends a notification to the user for a their booking before 6 hrs,3 hrs,1 hrs, 30 min and 15 minutes.
4. payment-service
   1. Payment service connects with the external systems and process a user payment.
   2. It maintains the payment details and the transaction details into a separate tables for the further reference and the tracing.
5. caching:
   1. To serve the read data faster and to reduce the database hits can use the caching.
   2. In the cache the latest and the trending events/movies which can be fetch faster.
   3. Also the seat availability of the frequently used event will also be cache so the user will get the seat availability details faster.
   4. Cache can also be use to reserve the seat when user processing for a booking request. this is the short live data in the cache.
6. CDN:
   1. CDN help to load the event/movie images faster also all static content can be load faster using CDN

Database Design

booking table: id, movieid, booking date, booking time, booking by, created datetime, updated datetime, status

transaction: id, externaltxnid, userid, datetime, bookingid, amount, paymentmode

Detailed Component Design

Caching:

1. Caching is one of the important here as the system is red heavy. following details will be store in side the catch.
2. Top events/movies: In this cache the top events/movies details will be store and its evection policy will be LFU.
3. Booking availability: here the seat availability details will be store. the key will be the event id and value will be available seat. The evection policy for this cache will be the LFU.
4. Reserve seat: This cache s a short leave cache which holder the Movie/Event id and seat ids as a key and user details as a value. its TTL will be 5 minuets.

Search Service:

1. This service get the request related to view the movies details, search details etc.
2. As the application is read heavy this service gets the more number of request and it can scale horizontally to handle the load.
3. When user loads the dashboard user will get the list of top 10 events/movies from the cache.
4. The latest event details will be retrieve from the cache and as a fallback it lookup into the database.
5. if user process with any of the movie/event for the booking its details first lookup into the cache if not found then it will be loaded from the database and put it into the cache before returning to the client.
6. The seat availability will also get loaded from the cache and if not exist then it pull from DB and load into the cache before returning to client.

booking service:

1. Once user select the seat and click on the book ticket. it makes the entry inside the cache to reserve the selected seat of a short time of period or till the time payment gets confirmed.
2. Once Cache entry made it sends the request to the payment service for a booking payment which further makes the payment and return the transaction number.
3. Once payment successfully completed it removes the entry from the reservation cache and put the entry into the the booking availability cache about the booked seat.
4. the booking event then publish to the kafka.

Notification service:

1. This service consumes and booking event and get the booking details and sends an notification to the customer by any communication medium such as Message, app notification, email etc.
2. This service contains multiple scheduled jobs which trigger after every 15 minutes and get details of the booking in past 1hours and sent the user notification about their upcoming booked event