👀 simultaneous-viewing-limit-system

Limit simultaneous viewing with Apache Kafka & Spring Boot Microservices, Redis

Overview

📲 About OTT Service..

Most people use media services. Recently, users (especially because of the covid-19 😷 ) have been staying indoors, and usage has been increasing. In particular, it uses OTT services, which have subscription-based pricing services.

Therefore, there is a trend that users share their accounts and use them together. Are there any policy or technical issues that could arise from this?

If it is an account sharing service, there is an issue about simultaneous viewing restrictions. Then, let's implement this simultaneous viewing limit asynchronously and expandable from the basics through various micro-services! 👩🏻‍💻👨‍💻

📮 Why Redis?

In release server, there will be a lot of connections, and watch logs.

This system checks log with key = set-top-box_id & hashkey = unique vod episode_id for searching faster

🧤 Why Kafka?

Micro services need to response each other async.

So, use Kafka as a Message Queue.

Produce topic, and Consume each topics that micro service need to subscribe.

📝 Before we start..

1. Push service ( Spring Boot Micro service )

2. Check service ( Spring Boot Micro service)

3. Redis Cluster

   • 3 master nodes ↔ 3 slave nodes

4. Kafka Cluster

   • 3 kafka cluster ↔ 1 zookeeper server (Actually, 3 zookeeper server is proper)

All of these servers are executed by localhost:portnumber

System Architecture

System Requirements

• Java 8
• Apache Kafka 2.3.0
• Redis 6.0.x

You can start Redis & Kafka cluster easily with these documents!

Redis Cluster tutorial

😎 Start Redis cluster easily

Kafka tutorial

🙄 Start Kafka easily

Server ports

Push Service : 9002
Check Service : 9000
Redis Cluster : 7001 ~ 7005
Kafka Cluster : 9093 ~ 9095
Zookeeper : 2182

Required Parameter

Request Params for Start VOD

{
  pcid = "username",
  episode_id = "episode_id",
  stb_id = "stb_id",
  play_start = "YYYY-mm-dd'T'HH:mm:ss",
  mac_address = "mac_address",
  running = true
  
}

Request Params for Stop VOD

{
  pcid = "username",
  episode_id = "episode_id",
  stb_id = "stb_id",
  play_start = "YYYY-mm-dd'T'HH:mm:ss",
  play_end = "YYYY-mm-dd'T'HH:mm:ss",
  mac_address = "mac_address",
  running = false
}

API Server

METHOD	PATH	DESCRIPTION
POST	:9002/send	Send request data for start watching VOD when new user connects
POST	:9002/force	Send request data for start watching VOD when another user exists
PATCH	:9002/stop	Send request data for stop watching VOD
POST	:9000/connect	Send request data for check connected user

PATH	DESCRIPTION
/queue/notify	send notify to user who subscribed each channel
/queue/disconnect	send disconnect notify to user who subscribed each channel
/queue/connect	send success response to user who subscribed each channel

Execute Spring boot application

1. Execute with mvnw

$ ./mvnw spring-boot:run

2. Maven Packaging

$ mvn package

3. Execute with jar file

$ java -jar target/push-service-0.0.1-SNAPSHOT.jar
$ java -jar target/check-service-0.0.1-SNAPSHOT.jar

Conclusion

Now you can use this Viewing-Limit-System.

Contact with my github profiles / dbals4818@gmail.com