[TODO] Long-term TODO list

[0oshowero0]

New Features

• [Roadmap] StreamingDataLoader for task-separated RL post-training TransferQueue/TransferQueue#85
• Implement check_data_consumption_status & check_data_production_status for TransferQueueController [Feat]: add check_data_production_status and check_consumption_status and support Polling get metadata TransferQueue/TransferQueue#157

Backend Related

• Support Ray RDT backend [Feat]: Add RayStorageClient to support the RDT feature of Ray TransferQueue/TransferQueue#167
• Support MoonCakeStore backend [Feat] Support Mooncake Store backend TransferQueue/TransferQueue#162

Adaption Works

• Support verl multi-modal scenario https://github.com/TransferQueue/verl/tree/agentloop_tq
• Merge with verl main code entry
• Support verl router replay https://github.com/TransferQueue/verl/tree/new_router_replay

Performance

• Optimize data spliting & gathering opterations (get_transfer_data, StorageUnitData.put_data/get_data, etc.) @jianjunzhong [Refactor] Simplify the put/get logic of AsyncSimpleStorageManager TransferQueue/TransferQueue#151
• Optimize the socket event loop by splitting into to multiple threads, one for each type of request
• Strengthen TransferQueueController concurrency
• Improve zero-copy performance [BREAKING][Perf] Optimize TransferQueue memory usage TransferQueue/TransferQueue#163 [4/4][serialization] Perf: support non-jagged nested tensor for zero-copy serialization in SimpleUnit backend TransferQueue/TransferQueue#121

User Interface

• Unified environmental variable management

Debug

• Performance metric [Perf] Add performance monitor tool & optimize log output TransferQueue/TransferQueue#132
• Command line debug interface
• Cleanup redundant logs [Misc] Clean up redundant logs & unify log format TransferQueue/TransferQueue#160 [BREAKING][Feat] Add fine-grained clear method & optimize log TransferQueue/TransferQueue#159 [Perf] Add performance monitor tool & optimize log output TransferQueue/TransferQueue#132
• Better metadata print [Feat] Add select_samples & improve print readability for BatchMeta TransferQueue/TransferQueue#133

Minor Issues & Code Clean

• Provide a general dynamic socket function for both Client & Storage
• Make dtypes & shapes compulsory for update_production_status
• Optimize _merge_tensors_to_tensordict to serve as a general function since similar codes may scatter everywhere inside TQ
• Split the zero copy optimization from zmq_utils.py to serial_utils.py providing a general serialization util for tensordict (both device-side and cpu-side) [Refactor] Provide a general zero-copy serial util TransferQueue/TransferQueue#140

Documentary

• Build doc web site
• Provide tutorials [Tutorial] Provide tutorials for TransferQueue usage TransferQueue/TransferQueue#141