chrislu
deb315a8a9
Phase E2: Integrate Protobuf descriptor parser with decoder - Update NewProtobufDecoder to use ProtobufDescriptorParser - Add findFirstMessageName helper for automatic message detection - Fix ParseBinaryDescriptor to return schema even on resolution failure - Add comprehensive tests for protobuf decoder integration - Improve error handling and caching behavior This enables proper binary descriptor parsing in the protobuf decoder, completing the integration between descriptor parsing and decoding. Phase E3: Complete Protobuf message descriptor resolution - Implement full protobuf descriptor resolution using protoreflect API - Add buildFileDescriptor and findMessageInFileDescriptor methods - Support nested message resolution with findNestedMessageDescriptor - Add proper mutex protection for thread-safe cache access - Update all test data to use proper field cardinality labels - Update test expectations to handle successful descriptor resolution - Enable full protobuf decoder creation from binary descriptors Phase E (Protobuf Support) is now complete: ✅ E1: Binary descriptor parsing ✅ E2: Decoder integration ✅ E3: Full message descriptor resolution Protobuf messages can now be fully parsed and decoded Phase F: Implement Kafka record batch compression support - Add comprehensive compression module supporting gzip/snappy/lz4/zstd - Implement RecordBatchParser with full compression and CRC validation - Support compression codec extraction from record batch attributes - Add compression/decompression for all major Kafka codecs - Integrate compression support into Produce and Fetch handlers - Add extensive unit tests for all compression codecs - Support round-trip compression/decompression with proper error handling - Add performance benchmarks for compression operations Key features: ✅ Gzip compression (ratio: 0.02) ✅ Snappy compression (ratio: 0.06, fastest) ✅ LZ4 compression (ratio: 0.02) ✅ Zstd compression (ratio: 0.01, best compression) ✅ CRC32 validation for record batch integrity ✅ Proper Kafka record batch format v2 parsing ✅ Backward compatibility with uncompressed records Phase F (Compression Handling) is now complete. Phase G: Implement advanced schema compatibility checking and migration - Add comprehensive SchemaEvolutionChecker with full compatibility rules - Support BACKWARD, FORWARD, FULL, and NONE compatibility levels - Implement Avro schema compatibility checking with field analysis - Add JSON Schema compatibility validation - Support Protobuf compatibility checking (simplified implementation) - Add type promotion rules (int->long, float->double, string<->bytes) - Integrate schema evolution into Manager with validation methods - Add schema evolution suggestions and migration guidance - Support schema compatibility validation before evolution - Add comprehensive unit tests for all compatibility scenarios Key features: ✅ BACKWARD compatibility: New schema can read old data ✅ FORWARD compatibility: Old schema can read new data ✅ FULL compatibility: Both backward and forward compatible ✅ Type promotion support for safe schema evolution ✅ Field addition/removal validation with default value checks ✅ Schema evolution suggestions for incompatible changes ✅ Integration with schema registry for validation workflows Phase G (Schema Evolution) is now complete. fmt
SeaweedMQ Message Queue on SeaweedFS (WIP, not ready)
What are the use cases it is designed for?
Message queues are like water pipes. Messages flow in the pipes to their destinations.
However, what if a flood comes? Of course, you can increase the number of partitions, add more brokers, restart, and watch the traffic level closely.
Sometimes the flood is expected. For example, backfill some old data in batch, and switch to online messages. You may want to ensure enough brokers to handle the data and reduce them later to cut cost.
SeaweedMQ is designed for use cases that need to:
- Receive and save large number of messages.
- Handle spike traffic automatically.
What is special about SeaweedMQ?
- Separate computation and storage nodes to scale independently.
- Unlimited storage space by adding volume servers.
- Unlimited message brokers to handle incoming messages.
- Offline messages can be operated as normal files.
- Scale up and down with auto split and merge message topics.
- Topics can automatically split into segments when traffic increases, and vice verse.
- Pass messages by reference instead of copying.
- Clients can optionally upload the messages first and just submit the references.
- Drastically reduce the broker load.
- Stateless brokers
- All brokers are equal. One broker is dynamically picked as the leader.
- Add brokers at any time.
- Allow rolling restart brokers or remove brokers at a pace.
Design
How it works?
Brokers are just computation nodes without storage. When a broker starts, it reports itself to masters. Among all the brokers, one of them will be selected as the leader by the masters.
A topic needs to define its partition key on its messages.
Messages for a topic are divided into segments. One segment can cover a range of partitions. A segment can be split into 2 segments, or 2 neighboring segments can be merged back to one segment.
During write time, the client will ask the broker leader for a few brokers to process the segment.
The broker leader will check whether the segment already has assigned the brokers. If not, select a few brokers based on their loads, save the selection into filer, and tell the client.
The client will write the messages for this segment to the selected brokers.
Failover
The broker leader does not contain any state. If it fails, the masters will select a different broker.
For a segment, if any one of the selected brokers is down, the remaining brokers should try to write received messages to the filer, and close the segment to the clients.
Then the clients should start a new segment. The masters should assign other healthy brokers to handle the new segment.
So any brokers can go down without losing data.
Auto Split or Merge
(The idea is learned from Pravega.)
The brokers should report its traffic load to the broker leader periodically.
If any segment has too much load, the broker leader will ask the brokers to tell the client to close current one and create two new segments.
If 2 neighboring segments have the combined load below average load per segment, the broker leader will ask the brokers to tell the client to close this 2 segments and create a new segment.