full-stack-microservice

The goal is to design a working microservice architecture with the following components:

Infrastructure

• Terraform

Load Balancer

• Nginx
• HAProxy
• Traefik
• Envoy
• Linkerd
• Fabio

Scheduler

• Nomad

Orchestration

• Docker Swarm
• Kubernetes

API Gateway

• Kong
• Tyk
• express-gateway
• API Umbrella

Service Mesh

• Linkerd
• Conduit
• Istio

Search

• Elasticsearch
• Solr
• Bleeve
• DNA Search

Cache

• Redis (also for rate-limiting, time-series analytics)
• Memcached
• Varnish
• Zipnish

Time-Series Database

• TimescaleDB
• InfluxDB (cluster is only available in Enterprise version though)

Tracing

• OpenTracing
• OpenZipkin
• Jaeger

Auth

• OpenId Connect
• OAuth2
• Identity Server 4

Secrets

• DockerSecrets
• Vault

Transport

• Kafka
• Nats
• gRPC
• protobuff
• Thrift
• Avro
• RabbitMQ
• ZeroMQ

Documentation

• Swagger

API Standards

• JSON Schema
• JSON API

Security

• letsencrypt

Others

• Client/server-side service discovery (nodejs resilient, etcd, Linkerd, Consul)
• Consul/etcd setup
• Circuit breaker
• Docker-compose or nomad setup
• Dynamic ports and service registry
• 12-Factor app practices
• Heteregenous clients (nodejs, go, python)
• Transport protocol (kafka, nats, rabbitmq, rpc, grpc, protobuff)
• Caching (Redis/memcached)
• API Gateway (kong, API Umbrella, AWS API Gateway)
• OpenID connect
• Sharding/Clustering of storage
• Vault
• RASA NLU
• Opentracing with GRPC
• GRPC security with Go, nodejs, python
• GRPC with graphql adapter
• Finanz example
• Create architecture diagram for all the different setups