When systems exchange customer events, they need an agreed structure that does not depend on either system’s programming language. Avro supplies a schema for that contract and a compact binary encoding for the data. The encoded data omits field names and relies on the schema to describe its structure. section 6.3.1
Listing 6.3 An example of the Avro format
{
"type": "record", #1
"name": "Client",
"namespace": "com.example", #2
"fields": [
{ "name": "clientId", "type": "string" }, #3
{ "name": "clientName", "type": "string" },
{ "name": "age", "type": "int" },
{ "name": "email", "type": ["null", "string"], "default": null }, #4
{ "name": "isPremium", "type": "boolean", "default": false },
{ "name": "address",
"type": {
"type": "record", #5
"name": "Address",
"fields": [
{ "name": "street", "type": "string" },
{ "name": "city", "type": "string" },
{ "name": "postalCode", "type": "string" }
]
}
},
{ "name": "createdAt", "type": "long", "logicalType": "timestamp-millis" }
]
}
The Client schema shows a record with a namespace, typed fields, a nullable email, a default value, a nested address, and a timestamp field. section 6.3.1
Applications can generate language-specific structures from an Avro schema or work dynamically with name–value pairs. Compared with a text-based JSON approach, Avro favors compact data and strong compatibility rules, but its binary messages are not human-readable. Its libraries and tools across popular languages also make it a practical choice for prototyping and troubleshooting. section 6.3.1 section 6.3.2
A data contract can be defined directly as an Avro schema, kept in Git, and referenced from an AsyncAPI specification. This avoids translating an abstract event model into a separate concrete schema. In Kafka deployments using Schema Registry, producers send a schema ID with the serialized data rather than sending the full schema in each message; consumers use that ID to retrieve the schema for deserialization. section 6.3.5 section 6.4.1
Kafka producers and consumers must be configured with matching Avro serializers or deserializers and a Schema Registry URL. In Kafka Connect, an AvroConverter instead translates between ConnectRecord objects and Avro-encoded Kafka data; the connector remains responsible for the external system’s native format. section 6.4.1 section 7.3.3
