Define read → transform → write as a single JSON job, and operate it like an engineering artifact
Describe read → transform → write as JSON so jobs are reusable, reviewable, and easy to operate.
Example job (swap reader/writer for your stack)
{
"job": {
"setting": {
"speed": { "channel": 4 }
},
"content": [{
"reader": {
"name": "mysqlreader",
"parameter": {
"username": "user",
"password": "******",
"connection": [{
"jdbcUrl": ["jdbc:mysql://host:3306/db"],
"table": ["source_table"]
}]
}
},
"writer": {
"name": "clickhousewriter",
"parameter": {
"username": "default",
"password": "******",
"connection": [{
"jdbcUrl": "jdbc:clickhouse://host:8123/db",
"table": ["target_table"]
}]
}
}
}]
}
}A large set of built-in plugins covers mainstream RDBMS, common NoSQL, and data-lake ecosystems.
