Please enable JavaScript.
Coggle requires JavaScript to display documents.
Microservices (Deployment (Multiple Service Instances per Host, Service…
Microservices
Deployment
Multiple Service Instances per Host
Service Instance per Host Pattern
Service Instance per Container Pattern
Format
RAML
Swagger
Inter process communication
AMQ
ZeroMQ
API Gateway
Tasks
Load balancing
Authentication
Caching
Aggregating results
Service Discovery
Client side Discovery Pattern
Server Side Discovery Pattern
Centralized discovery logic
register('ip', 'servicename')
unregister()
heartbeat()
Load balancer directs calls to correct service instance
Tools
Apache Zookeeper
consul.io
each microservice has its own private datastore
Maintaining data consistency across services and databases is essential,
event‑driven architecture.
BASE model <> ACID model
materialized views
CAP theorem
achieving atomicity
Event Sourcing
(+) reliably publish events whenever state changes
(+) solves data consistency issues
(+) avoids the object‑relational impedance mismatch problem
(+) 100% reliable audit log
(-) different and unfamiliar style of programming
CQRS
Local Transactions
(+) guarantees an event is published
(-) developer must remember to publish events
Transaction log mining
(+) guarantees that an event is published
(+) Separating event publishing from the application’s business logic
(-) format of the transaction log is proprietary to each database