zbtrs2
/
Mooncake
forked from mooncake-track/Mooncake
0
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Commit Graph

  • ca8674f51b test(store): normalize allocator test headings feature/ollama-kvcache-bus zbtrs2 2026-06-29 22:29:34 +0800
  • fd7bbf720a ollama-kvcache: add deployment, observability and docs zbtrs2 2026-06-29 01:52:29 +0800
  • 897a11bd34 ollama-kvcache: add benchmarks and orchestration scripts zbtrs2 2026-06-29 01:52:12 +0800
  • e31c2de9b8 ollama-kvcache: add Ollama integration patches zbtrs2 2026-06-29 01:51:49 +0800
  • 0b18e65193 ollama-kvcache: add Stage-2 on-device KV state path zbtrs2 2026-06-29 01:51:33 +0800
  • 56efccf73c ollama-kvcache: add Mooncake store proxy zbtrs2 2026-06-29 01:51:08 +0800
  • 0b208be4e6 ollama-kvcache: add sidecar orchestrator, llama.cpp client and server zbtrs2 2026-06-29 01:50:44 +0800
  • 4927b5a694 ollama-kvcache: add gRPC contracts and store backends zbtrs2 2026-06-29 01:50:25 +0800
  • 2febfc17e6 ollama-kvcache: add cache-key, prefix index and cost arbiter zbtrs2 2026-06-29 01:50:09 +0800
  • ddcc56c2f6 [Doc] ollama integration: showcase README, report figures, demo scoreboard feat/ollama-kvcache-integration zbtrs2 2026-06-24 23:10:45 +0800
  • 3be953c8fa [Doc] ollama integration: cohesive figure suite for the report zbtrs2 2026-06-24 23:10:28 +0800
  • 7612314050 [Doc] ollama integration: report, figures and docs wiring zbtrs2 2026-06-23 19:10:16 +0800
  • 935b5880d0 [Integration] ollama: additive, env-gated patches for ollama/ollama zbtrs2 2026-06-23 19:10:16 +0800
  • f71f67a8a5 [Integration] ollama: deployment configs and workspace scripts zbtrs2 2026-06-23 19:10:16 +0800
  • e1588559bd [Integration] ollama: multi-agent benchmark harness zbtrs2 2026-06-23 19:10:16 +0800
  • ce62f4b670 [Integration] ollama: Stage-2 KV-state microbenchmark (libllama) zbtrs2 2026-06-23 19:10:16 +0800
  • b678593892 [Integration] ollama: Python store-proxy over MooncakeDistributedStore zbtrs2 2026-06-23 19:10:16 +0800
  • 899eef2cdf [Integration] ollama: sidecar daemon, gRPC/HTTP servers, Prometheus metrics zbtrs2 2026-06-23 19:10:16 +0800
  • cef91c4dcd [Integration] ollama: three-stage reuse orchestrator zbtrs2 2026-06-23 19:10:16 +0800
  • 113076bc96 [Integration] ollama: llama.cpp bridge (Stage-1 /slots, Stage-2 cgo) zbtrs2 2026-06-23 19:10:16 +0800
  • a2d8cc30fb [Integration] ollama: pluggable KV-snapshot store backend zbtrs2 2026-06-23 19:10:16 +0800
  • cebbcb20e6 [Integration] ollama: restore-vs-recompute cost arbiter zbtrs2 2026-06-23 19:10:16 +0800
  • 99781dcafa [Integration] ollama: cross-process radix index of KV prefixes zbtrs2 2026-06-23 19:10:16 +0800
  • 3abd2419e0 [Integration] ollama: content-addressed cache keys with chained block hashing zbtrs2 2026-06-23 19:10:16 +0800
  • 83ba81669e [Integration] ollama: gRPC contracts for the KV-cache bus zbtrs2 2026-06-23 19:10:16 +0800
  • 036075ccfa
         [Bugfix]: route cache allocator capacity adapt (#1958) P2P-Mooncake-Store Shelron 2026-04-23 12:50:19 +0800
  • aad5111781
         [Store] Support degraded startup when master is unavailable (#1930) skygragon 2026-04-22 20:56:56 +0800
  • 188bb41121
         [Store] Remove read pool and slot of stress test (#1955) Wan 2026-04-22 19:09:51 +0800
  • 24e29df083
         [Store][K8s-Native][1/N] Go-library for k8s leader election (#1910) main Vladislav Nosivskoy 2026-04-22 12:53:16 +0300
  • db7bce3056
         [Store] Support storage hierarchy with offload-on-evict mode (#1899) Geonwoo Choi 2026-04-22 18:48:07 +0900
  • 29e2c7e873
         [Store] optimize stress test (#1953) Wan 2026-04-22 16:32:01 +0800
  • defbadd6a5
         fix(store-rust): remove duplicated smoke test and benchmark blocks add_rust_test2 copilot-swe-agent[bot] 2026-04-22 04:55:49 +0000
  • a14e0b600a
         [Store] fix: setup logs dir for realclient_main (#1925) 人生若只如初见 2026-04-22 12:54:51 +0800
  • 7442626169
         [Build] Update build_wheel.sh package installation logic (#1950) Shangming Cai 2026-04-22 12:49:47 +0800
  • b888a61f4c
         Update build_wheel.sh package installation logic update-build-1 Shangming Cai 2026-04-22 11:40:57 +0800
  • a6bfb2e195 do not use lru list defautly & add log for remove-event dev/kv-indexer yejj710 2026-04-22 11:34:22 +0800
  • e1d6d6f6f4
         Bump version to 0.3.10.post2 in pyproject.toml (#1949) v0.3.10.post2 Shangming Cai 2026-04-22 11:01:31 +0800
  • 5e701e15aa
         Bump version to 0.3.10.post2 in pyproject.toml bump-patch-1 Shangming Cai 2026-04-22 11:00:15 +0800
  • 255e287bc1
         Refactor ASIO shared target into mooncake-common (#1926) Copilot 2026-04-22 00:15:33 +0800
  • cfea2cb0f5
         [Common][Etcd] Pin a patched Go toolchain and track etcd go.sum for wrapper builds (#1937) Copilot 2026-04-21 22:39:48 +0800
  • 0d96064ee0
         [Store] Optimize batch_put with BatchGetWriteRoute RPC (#1947) Wan 2026-04-21 22:28:25 +0800
  • 9408d853d6 modify asio target dir copilot/move-mooncake-asio-common zhang zuoyuan 2026-04-21 20:53:40 +0800
  • 23b5886e8c fix compile err zhang zuoyuan 2026-04-21 20:04:20 +0800
  • 67b3f9de2e refactor: move asio shared target into mooncake-common copilot-swe-agent[bot] 2026-04-18 16:31:11 +0000
  • ee69b7ea40 Initial plan copilot-swe-agent[bot] 2026-04-18 16:19:29 +0000
  • 8a61d5b47c
         [Store]: Wait for all tasks to complete in TransferEngineOperationState::check_task_status (#1906) Jinlong Chen 2026-04-21 21:09:43 +0800
  • 741cf0adff
         [PG] Fix wait() hang during CUDA Graph capture (#1933) Zhanhao Cao 2026-04-21 18:24:23 +0800
  • b21f90d032
         [Store] Add HTTP metrics endpoint for Client (#1934) skygragon 2026-04-21 17:35:20 +0800
  • 1e9fa36703
         [CI] Restore auto-triggered ascend-test and integration-test in ci.yml (#1943) lujh 2026-04-21 15:58:01 +0800
  • 5078873532
         fix race (#1932) Ruoyu Qin 2026-04-21 15:07:21 +0800
  • 38c3975138
         [Store] auto-enable MC_STORE_MEMCPY in TCP-only environments (#1936) dtc 2026-04-21 15:01:46 +0800
  • f9dd50c543
         [Integration] connector_v1: subclass SupportsHMA so PD-disagg works for hybrid models (#1931) Harshavardhan K 2026-04-20 23:31:28 -0700
  • 125ac89379
         Track etcd go.sum for wrapper builds copilot/fix-vulnerable-go-toolchain copilot-swe-agent[bot] 2026-04-21 04:26:22 +0000
  • 0f22234d0b
         [Store] unify file storage backend env vars under MOONCAKE_OFFLOAD_ p… (#1929) copilot/compare-cudamemcpy-cu12-cu13 lujh 2026-04-21 11:36:54 +0800
  • c58d1f90b9
         [store] Bug Fix: Local Disk Replica Metadata Not Cleaned Up After Store Node Offline (#1914) Colors-111 2026-04-21 11:14:45 +0800
  • b9de4a9892
         fix: add mooncake a2a backend flag to moe tone test copilot/fix-e2e-issue-mooncake copilot-swe-agent[bot] 2026-04-21 03:11:03 +0000
  • ac53c874ba
         [Store] Add client bandwidth metrics for real and dummy clients (#1874) Teng Ma 2026-04-21 11:06:52 +0800
  • bb5404c7ca
         Pin patched Go toolchain for etcd wrapper builds copilot-swe-agent[bot] 2026-04-21 03:02:27 +0000
  • 414440316f
         Initial plan copilot-swe-agent[bot] 2026-04-21 02:58:38 +0000
  • 952da65651
         [TE] PTE-aware auto-split large MR registration for EFA transport (#1912) 王鹤男 2026-04-20 15:04:49 +0800
  • cd67a36da0
         [TENT] add FaultProxyTransport for fault injection testing (#1907) Stary 2026-04-20 14:45:47 +0800
  • 1fbe35c2fe
         fix(ci): retry ascend submodule update via GitHub mirrors (#1924) Stary 2026-04-20 14:32:39 +0800
  • 3e7c78de9b
         [Store] Fix segfault in disk-replica/offload paths when handling GPU pointers (#1892) lujh 2026-04-20 11:51:53 +0800
  • 90f2005b55
         [Store] Parallelize Local and Remote Read/Write in Batch Scenarios (#1921) Wan 2026-04-20 11:47:43 +0800
  • 62651b325b
         [Bugfix] Minor bug fixes and improvments for Mooncake Store and Transfer Engine (#1895) Jinlong Chen 2026-04-20 00:28:00 +0800
  • e985058d08 [Store] Add Rust smoke test, benchmark, and CI coverage Teng Ma 2026-04-19 02:12:47 +0800
  • 32329c8356
         [TENT] Batch transfer requests using cudaMemcpyBatchAsync (#1890) shuoerw 2026-04-20 00:06:40 +0800
  • c837624931 Add LRU eviction policy to the index data of Mooncake-Store yejj710 2026-04-19 14:32:54 +0800
  • 7d443489c2
         [CI] add format hook (#1904) Teng Ma 2026-04-18 01:31:24 +0800
  • 6fd613e370
         [CI] Improve code format check to only scan changed files (#1922) skygragon 2026-04-17 23:04:04 +0800
  • b0214baf0e
         [Test] Fix HA integration test flakiness (#1916) skygragon 2026-04-17 18:10:06 +0800
  • 28464f3aee
         [Store] Expose batch_replica_clear in Python binding (#1848) Geonwoo Choi 2026-04-17 18:36:32 +0900
  • 84a823b9eb
         [Build] Add memory-aware compile/link parallelism with Ninja support (#1909) skygragon 2026-04-17 10:52:14 +0800
  • e878cb2312
         [Store] Enabling setting SSD offload path using python interface (#1884) zhangzuo21 2026-04-17 10:47:57 +0800
  • f750e9d8ab
         [Doc] Update CODEOWNERS with new document owners stmatengss-patch-4 Teng Ma 2026-04-17 10:07:50 +0800
  • a6cbc1a417
         [TE] Add fi_read support, endpoint LRU eviction, and multi-NIC striping for EFA transport (#1821) 王鹤男 2026-04-17 01:30:39 +0800
  • 67bd66c2f6
         docs: implement documentation plan for missing Mooncake features copilot/list-lacking-documents copilot-swe-agent[bot] 2026-04-16 16:52:56 +0000
  • 9ce5d25292
         update docs (#1905) ykwd 2026-04-16 23:23:11 +0800
  • 0a9c9937c7
         [PG][TENT] Fix CUDA collective wait semantics and NVLink small-transfer completion (#1863) ZiWei Yuan 2026-04-16 20:39:28 +0800
  • dc965d3121
         [TENT] Fix static library link group for final targets and reformat related CMake files (#1893) Stary 2026-04-16 20:28:25 +0800
  • 8d3beecb28
         [Store] Fix hardcoded 127.0.0.1 bind address in standalone client RPC (#1900) zhangzuo21 2026-04-16 20:20:37 +0800
  • 15d99a3002
         feat(tent): replace raw RdmaEndPoint* with weak_ptr for endpoint lifecycle safety (#1897) Stary 2026-04-16 20:13:33 +0800
  • fb3ffcac10 [CI] add format hook add_format_hook Teng Ma 2026-04-16 19:04:21 +0800
  • 020942121d
         [CI] add configurable GitHub mirror fallback for Ascend checkout (#1896) Stary 2026-04-16 14:57:03 +0800
  • 4a8684bc3e
         [CI] fix: health_check_test flaky due to fixed sleep, use polling for master down detection (#1868) Skyper 2026-04-16 14:36:32 +0800
  • c44a052944
         [Store] Enable NUMA-segmented allocation for all RDMA deployments (#1838) fatSheep 2026-04-15 17:27:57 +0800
  • d85f9bd82f
         [TENT] Fix crashes caused by negative numa_node and incorrect config type inference (#1894) dtc 2026-04-15 17:15:38 +0800
  • c2573405bb
         [Store] Support SSD Metrics (#1879) lujh 2026-04-15 12:01:18 +0800
  • 041ddb1794
         [TENT] Enhance memory registration with transport type support (#1877) Stary 2026-04-15 10:00:32 +0800
  • aadcc7daf1
         [TENT] Wire up cross-transport failover with safety limits and observability (#1878) Stary 2026-04-15 09:52:41 +0800
  • cfc5cf4c2a
         [TransferEngine][MACA] Align MACA build paths with CMake options (#1888) Dayuxiaoshui 2026-04-15 09:47:39 +0800
  • 3a69fa4b4d
         [TE] Enabling UB Transport on the Kunpeng SuperNode Phase 2 (#1855) Chuang Zhang 2026-04-15 00:15:34 +0800
  • 01900be50b
         [MISC] Add CODEOWNERS for efa_transport directory (#1885) Teng Ma 2026-04-14 16:37:19 +0800
  • fa0641a436
         [MISC] Add CODEOWNERS for efa_transport directory stmatengss-patch-3 Teng Ma 2026-04-14 14:20:05 +0800
  • a175e9aba8
         [Store] Support SSD offload via Python setup() interface (#1857) zhangzuo21 2026-04-14 11:48:34 +0800
  • c67649c83a
         support remove event & add some info-level log (#1881) yejj 2026-04-13 21:09:18 +0800
  • 6ab0016ed3
         [Store ]Implement Master HA recovery based on client (#1876) Wan 2026-04-13 19:33:23 +0800
  • 738b375338
         [Docs] Add SSD offload benchmark results (#1835) zhangzuo21 2026-04-13 15:52:52 +0800
  • c0d07af568
         [TransferEngine][ROCm] Add ROCm HIP support to the Mooncake Python package (#1742) Zejian Wang 2026-04-13 11:13:10 +0800
  • 1a30e4e110
         [TransferEngine] Add retry, async execution, and graceful shutdown for TENT TCP transport (#1866) Stary 2026-04-13 09:58:11 +0800