Current state: Accepted
Released: with Milvus 2.1
The log broker is a pub-sub system within Milvus, It is responsible for streaming data persistence, event notification, recovery etc. Now Milvus cluster mode uses Pulsar as a log broker, and standalone mode uses RocksDB.
Apache Kafka is a distributed event store and stream-processing platform, and it is a popular solution for data streaming needs. Many community users expect Milvus to support Kafka because they have already used it in the production environment.
Milvus supports Kafka as a message stream, we can use the configuration option to decide to use Pulsar or Kafka on cluster mode.
If we want to enable Kafka, need to config it in milvus.yaml file, Pulsar is also. when Kafka and Pulsar config at the same time, Pulsar has higher priority use.
Kafka Client SDK
We tried using sarama and confluent-kafka-go in our development and found that there was basically no difference in the producer. But there is a big difference when using consumer groups.
Sarama uses consumer group needs to implement Sarama interfaces. it is very difficult to control and hard to seek.
confulent-kafka-go use consumer group to consume messages just a function. It is very simple to use. This function allows you to directly set the offset from which to start consumption.
Note: in order to provide a unified MQ interface, We will remove a series of Reader API from MsgSream interface, actually Kafka does not support Reader API and it also is implemented by Consumer API.
Implement Kafka consumer based on the following interface:
Implement Kafka producer based on the following interface:
on the other hand, we must provide a configurable ability for MQ clients, so SetParams API will be replaced by Init API.
- Helm Chart
- pass the ut
- performance testing
- chaos testing