Multi-cloud deployment

When you need to be able to run anywhere

Project overview

CLIENT

Proximetry

INDUSTRY

IoT

LOCATION

Katowice, Poland

TIMELINE

10/2020 - 06/2021

Task

The Solution

The problem was quite complex because the number of components and deployment scenarios. We chose a scenario in which we run all system components in Kubernetes cluster.

Technology used

aws
azure
erlang
java
nodejs
cassandra
elasticsearch
rabbitmq
kafka
riak
postgresql
kubernetes

Infrastructure code

This is another, great example of building blocks approach usability. Ansible and Terraform modules mixture allowed us to create a core of the infrastructure which can be deployed on cloud and on-premise environments.

Kubernetes

To make sure that we use the same service always and everywhere, we opted for the self-hosted Kubernetes. This step allowed us to standardize the deployment and maintenance of both applications and services.

Services & microservices

All services, such as SQL and NoSQL databases, message brokers and microservices that make up the application have been containerized. The procedures related to building, testing and delivering components to the environment have been wrapped into pipelines.

Monitoring

The monitoring system was based on open-source solutions. For maintenance purposes, we have created a set of navigation tools and dashboards that allowed you to easily check the general condition of the system and thoroughly analyze each of the components.

The Result