Skip to content

Latest commit

 

History

History
98 lines (66 loc) · 2.42 KB

README.md

File metadata and controls

98 lines (66 loc) · 2.42 KB
Raw

ParkingLot

Packages

com.com.campaignmonitor.parkingLot.model

This package contains the POJO classes

com.com.campaignmonitor.parkingLot.service

This package contains the service layer classes The service implementations can use the following for database access and persistence.

  1. Spring JDBC template
  2. JPA
  3. Hibernate

com.com.campaignmonitor.parkingLot.controller

This package contains MVC Controller classes. The can be Restful API microservices implementations.

Testing

The model package

These classes can be tested with a generic model bean tester.

The service layer

These will contain unit test cases for your service layer classes.
I have stubbed possible testcases in each class.

The controller

The controllers can be tested using a BDD test framerwork (Cucumber).

Deployment

Hosting

AWS Web hosting/cloud.
This solution is reliable, scalable

OpenShift Web hosting/cloud.
This solution is reliable, scalable

Google Cloud Platform.
This solution is reliable, scalable
 
Oracle Cloud.
This solution is reliable, scalable

IBM Cloud.
This solution is reliable, scalable

Azure Web hosting/cloud.
This solution is reliable, scalable

On-premise
This approach is a lot more hands-on and not as reliable or scalable. Reliability and scalability
can be achieved but at higher costs.

Database

Oracle, Mysql

Deployment tools

Jenkins, Kebernetes, Docker, Nexus, Git, BitBucket

Clients

You can use a variety of different clients

Mobile

Desktop Browser

IoT embedded devices

Scaling

Database

You can use a traditional relational database as a MVP. If the data becomes too large, you can consider partitioning data, and also a nosql solution.

Messaging

As a MVP I choose a simple Restful api. You can consider a Kafka/NIFI solution. In a Kafka/NIFI solution, the "space" embedded/IOT devices can be producers that will publish messages to a "space" topic. You can then use apache NIFI to consume the messages from the Kafka "space" topic.

Bottlenecks

Database

TPS

Network

Network traffic

Webserver

You can horizontally scale by using a load balancer and adding more servers.