morning-commute

an event-driven ride-hailing simulation built with java, spring boot, and kafka.

flow

1. request: a rider creates a trip via the trips-service.
2. events: a TripRequested event is pushed to kafka.
3. pricing: the pricing-service calculates zone demand in real-time.
4. analytics: the analytics-service aggregates demand metrics into 1-hour windows and saves them to postgres.

tech

• java 21 & spring boot 3.5
• kafka & kafka streams
• postgres for olatp and analytics
• docker compose for local infra

services

• trips-service: rest api for trip lifecycle (create, complete, retrieve).
• pricing-service: calculates demand multipliers based on active trips.
• analytics-service: computes hourly zone metrics using kafka streams.

setup & run

build the jars:

./mvnw clean package -DskipTests -f common/pom.xml
./mvnw clean package -DskipTests -f services/trips-service/pom.xml
./mvnw clean package -DskipTests -f services/pricing-service/pom.xml
./mvnw clean package -DskipTests -f services/analytics-service/pom.xml

spin up the infrastructure:

cd infra && docker compose up -d --build

verify status:

curl http://localhost:8080/actuator/health
curl http://localhost:8081/actuator/health
curl http://localhost:8082/actuator/health

manual testing

trip lifecycle

create a trip:

curl -X POST http://localhost:8080/trips \
  -H "Content-Type: application/json" \
  -d '{"riderID": "rider-123", "pickupZone": "Z12", "dropoffZone": "Z05"}'

check status:

curl http://localhost:8080/trips/{trip-id}

complete trip:

curl -X POST http://localhost:8080/trips/{trip-id}/complete

query services

get pricing for a zone:

curl http://localhost:8081/pricing/zone/Z12

view aggregated analytics:

docker exec -it postgres psql -U user -d rides_db \
  -c "SELECT * FROM analytics_zone_hourly ORDER BY hour_bucket DESC LIMIT 10;"

troubleshooting

• ensure ports 8080-8082, 5432, 9092 are free.
• logs: docker compose logs -f <service-name>
• restart: docker compose restart <service-name>