pulm.src

@startuml
participant StackOverflow
actor Golang_ETL #red
participant HBase
participant S3
actor Cruncher #red
participant HDFS
actor WebRunner #red
group ETL
StackOverflow -> Golang_ETL: RSS Feed: Skills & Job Descs
Golang_ETL -> Golang_ETL: Extract Skills\n& Job URLs 
Golang_ETL -> HBase: Store Skills
StackOverflow -> Golang_ETL: Pull Jobs Descriptions
Golang_ETL -> Golang_ETL: Clean Jobs HTML
Golang_ETL -> S3: Store Cleaned Jobs
end
group Crunch Model
S3 -> Cruncher: Pull Jobs
HBase -> Cruncher: Pull Skills
Cruncher -> Cruncher: Create TF-IDF Model
Cruncher -> HDFS: Store TF-IDF Model 
end
group Web Interaction
HDFS -> WebRunner: Load Model
HBase -> WebRunner: Load Skills
actor WebUser #yellow
WebUser -> WebRunner: Resume Provided
WebRunner -> WebRunner: Calculate Jobs
WebRunner -> WebUser: Here are Jobs
WebUser -> StackOverflow: Review Jobs
end
@enduml