Plantuml

The AWS shapes are included in the PlantUML stdlib .. simply include them as shown here: example.puml @startuml !include <awslib/AWSCommon> !include <awslib/Analytics/ManagedStreamingforKafka> !include <awslib/Database/RDS> !include <awslib/General/Users> !include <awslib/General/InternetGateway> !include <kubernetes/k8s-sprites-labeled-25pct> skinparam linetype ortho title "AWS Context Diagram" package "EKS Kubernetes Cluster" as eks_cluster { component "<$pod>\napp1" as pod1 component "<$pod>\napp2" as pod2 } ManagedStreamingforKafka(kafka_pod, "Amazon MSK", "Apache Kafka") RDS(pg_rds, "PostgreSQL", "Sample Schema") Users(users, "AppUsers","editors, admins") InternetGateway(igw1, "Customer Gateway", "Customer access to internal services") users -> igw1 igw1 --> pod1 igw1 --> pod2 pod1 --> pg_rds pod1 --> kafka_pod pod2 --> pg_rds pod2 --> kafka_pod @enduml ...

C4 models allow us to visualize software architecture by decomposition in containers and components. Viewpoints are organized in hierarchical levels: Context Diagrams (Level 1) Container Diagrams (Level 2) Component Diagrams (Level 3) Code Diagrams (Level 4) C4-PlantUML offers a variety of macros and stereotypes that make modeling fun. An example in PlantUML: sample.puml @startuml !include <c4/C4_Context.puml> !include <c4/C4_Container.puml> left to right direction Person(user, "User") System_Ext(auth, "AuthService", "Provides authentication and authorization via OIDC") System_Boundary(zone1, "Some system boundary") { System(lb, "Load Balancer") System_Boundary(az, "App Cluster") { System(app, "App Servers") { Container(app1, "App1", "Docker", "Does stuff") Container(app2, "App1", "Docker", "Does stuff") ContainerDb(dbSess, "Session DB", "Redis") ContainerDb(db1, "RBMS 1", "AWS RDS Postgres") ContainerDb(db2, "RBMS 2", "AWS RDS Postgres") ' both app servers sync sessions via redis Rel(app1, dbSess, "Uses", "Sync Session") Rel(app2, dbSess, "Uses", "Sync Session") ' both app servers persist data in RDBMS Rel(app1, db1, "Uses", "Persist/query relational data") Rel(app2, db2, "Uses", "Persist/query relational data") } } } Rel(user, lb, "call") Rel(lb, app1, "delegate") Rel(lb, app2, "delegate") Rel(app1, auth, "Verify", "User auth") Rel(app2, auth, "Verify", "User auth") SHOW_FLOATING_LEGEND() @enduml ...

The following hack allows us to draw a diagram in a diagram in a table in a legend :D example-diagram.puml @startuml skinparam legendBackgroundColor transparent skinparam legendBorderColor transparent actor Editor as editor queue Queue as "Message\nQueue" editor -> [Component] : call [Component] --> Queue : publish legend right {{ scale 0.7 skinparam defaultFontSize 14 skinparam BackGroundColor transparent skinparam defaultBackgroundColor white !procedure $entry($type, $label, $scale=1) {{\nscale $scale \nskinparam backgroundcolor transparent\nlabel " " as A\nlabel " " as B\n $type \n}} => $label !endprocedure map "<b>Legend</b>" as legend #white { $entry(":Editor:","\nEditor for Data xyz", 0.6) $entry("[Component]","\nSome Component", 0.8) $entry("queue Queue","\nKafka-Topic", 1) $entry("A -> B","Data-Flow") } }} endlegend @enduml ...

Reactive Streams is an initiative trying to standardize asynchronous stream processing with non-blocking back-pressure. With Java 9, new classes in the java.util.concurrent.flow package offer a semantically equivalent counterpart to this standard that may be adopted by other frameworks. In the following short tutorial we’re implementing examples for reactive streams with Java 9 and the Flow API, with RxJava2, with Akka, with Reactor and finally there is an example in RxJava1, too though it does not follow the standard. ...