Mooncake Files

What's Changed

• Fix/tent batch transfer merge boundary by @Primary33 in https://github.com/kvcache-ai/Mooncake/pull/1704
• docs: add TorchSpec in Mooncake README by @zhyncs in https://github.com/kvcache-ai/Mooncake/pull/1709
• [Store] Add eviction policy for BucketStorageBackend with batch master notification by @zhangzuo21 in https://github.com/kvcache-ai/Mooncake/pull/1646
• Support duration units for master TTL flags by @Primary33 in https://github.com/kvcache-ai/Mooncake/pull/1684
• E2E storage backend test by @maheshrbapatu in https://github.com/kvcache-ai/Mooncake/pull/1660
• [CI] Add hixl roce samples on ASCEND platforms. by @VNightMare in https://github.com/kvcache-ai/Mooncake/pull/1697
• Remove non-portable GCC-internal headers from Transfer Engine by @Copilot in https://github.com/kvcache-ai/Mooncake/pull/1716
• [STORE] introduce HA backend abstraction by @YiXR in https://github.com/kvcache-ai/Mooncake/pull/1678
• [TENT] fix: avoid resetting RDMA endpoint on duplicate concurrent bootstrap by @00fish0 in https://github.com/kvcache-ai/Mooncake/pull/1705
• [Bugfix] Fix tent_metrics build errors with TENT_METRICS_ENABLED=ON by @staryxchen in https://github.com/kvcache-ai/Mooncake/pull/1712
• [PG] Fix group size expansion by @caozhanhao in https://github.com/kvcache-ai/Mooncake/pull/1706
• build: add memory-aware compile/link parallelism by @staryxchen in https://github.com/kvcache-ai/Mooncake/pull/1718
• [Skill] feat: add trouble shooting skill by @stmatengss in https://github.com/kvcache-ai/Mooncake/pull/1724
• [EP] make num_ranks more flexible by @ympcMark in https://github.com/kvcache-ai/Mooncake/pull/1725
• [Misc] Improve developers' experience for EP & PG by @UNIDY2002 in https://github.com/kvcache-ai/Mooncake/pull/1708
• [TE] refactor ascend direct transport & adapt to dummy real mode of store by @ascend-direct-dev in https://github.com/kvcache-ai/Mooncake/pull/1720
• [STORE] add Redis leadership backend and HA regression coverage by @YiXR in https://github.com/kvcache-ai/Mooncake/pull/1722
• store: split client HA/control-plane threads and suppress zero-seg he… by @XucSh in https://github.com/kvcache-ai/Mooncake/pull/1736
• [Transfer Engine] add initial MACA build path and CUDA-like adapter by @Dayuxiaoshui in https://github.com/kvcache-ai/Mooncake/pull/1731
• add required library when build by @xleoken in https://github.com/kvcache-ai/Mooncake/pull/1674
• [PG] optimize p2p-proxy buffer size by @JunlinW113 in https://github.com/kvcache-ai/Mooncake/pull/1735
• [Store] Add hard pin mechanism for eviction-protected objects by @he-yufeng in https://github.com/kvcache-ai/Mooncake/pull/1728
• [PG] Increase kP2PBufferSize to unlock full performance potential by @UNIDY2002 in https://github.com/kvcache-ai/Mooncake/pull/1740
• Fix/tent store metadata override by @Primary33 in https://github.com/kvcache-ai/Mooncake/pull/1743
• [bug fix]Inconsistent parameter of NVMeoFTransport::submitTransferTask by @yz53665 in https://github.com/kvcache-ai/Mooncake/pull/1748
• add script for metadata management by @whybeyoung in https://github.com/kvcache-ai/Mooncake/pull/1746
• [Store] adapt to dummy real mode for ascend by @ascend-direct-dev in https://github.com/kvcache-ai/Mooncake/pull/1723
• [STORE] abstract snapshot catalog and add Redis snapshot backend by @YiXR in https://github.com/kvcache-ai/Mooncake/pull/1739
• Fix/tent slice queue race by @Primary33 in https://github.com/kvcache-ai/Mooncake/pull/1737
• [PG] Add GPU barrier support for mooncake-pg by @LuLuji04 in https://github.com/kvcache-ai/Mooncake/pull/1751
• [STORE] support Redis ACL username authentication and reorganize HA by @YiXR in https://github.com/kvcache-ai/Mooncake/pull/1757
• [TENT] fix tebench RPATH to find libasio.so at runtime by @alogfans in https://github.com/kvcache-ai/Mooncake/pull/1750
• add missing steps to efa transport readme by @bob-bins in https://github.com/kvcache-ai/Mooncake/pull/1759
• [PG] Enable asynchronous recovered-rank initialization with deferred join by @UNIDY2002 in https://github.com/kvcache-ai/Mooncake/pull/1744
• [DOC] Add troubleshooting for RDMA MKEY resource exhaustion by @00fish0 in https://github.com/kvcache-ai/Mooncake/pull/1765
• [TE] Fixed an issue where start_timestamp is retrieved but batch_desc has already been freed by @hjchen2 in https://github.com/kvcache-ai/Mooncake/pull/1760
• [TE] Fix simultaneous open handshake in RdmaEndpoint by @caozhanhao in https://github.com/kvcache-ai/Mooncake/pull/1733
• [tent] Set Auto_Connect as default by @Cheng-China in https://github.com/kvcache-ai/Mooncake/pull/1758
• [PG][TENT] Fix first-collective hangs on NVLink/MNNVL bootstrap by @KMSorSMS in https://github.com/kvcache-ai/Mooncake/pull/1755
• fix(tent): address build issues and enable CI coverage by @Primary33 in https://github.com/kvcache-ai/Mooncake/pull/1768
• [Skill] run mooncake unit tests locally by @stmatengss in https://github.com/kvcache-ai/Mooncake/pull/1774
• [Store] batch remove by @LujhCoconut in https://github.com/kvcache-ai/Mooncake/pull/1756
• fix: correct TP zero-copy put semantics by @zxpdemonio in https://github.com/kvcache-ai/Mooncake/pull/1685
• [Store] Fix SSD offload failure in Metadata Server mode (#1729) by @zhangzuo21 in https://github.com/kvcache-ai/Mooncake/pull/1771
• Fix TENT Python binding initialization and build output by @XucSh in https://github.com/kvcache-ai/Mooncake/pull/1778
• Bump version to 0.3.10.post1 in pyproject.toml by @ShangmingCai in https://github.com/kvcache-ai/Mooncake/pull/1788

New Contributors

• @Dayuxiaoshui made their first contribution in https://github.com/kvcache-ai/Mooncake/pull/1731
• @JunlinW113 made their first contribution in https://github.com/kvcache-ai/Mooncake/pull/1735
• @yz53665 made their first contribution in https://github.com/kvcache-ai/Mooncake/pull/1748
• @LuLuji04 made their first contribution in https://github.com/kvcache-ai/Mooncake/pull/1751
• @bob-bins made their first contribution in https://github.com/kvcache-ai/Mooncake/pull/1759
• @LujhCoconut made their first contribution in https://github.com/kvcache-ai/Mooncake/pull/1756

Full Changelog: https://github.com/kvcache-ai/Mooncake/compare/v0.3.10...v0.3.10.post1