github.com/koordinator-sh/koordinator

What's Changed

• webhook: fix pod validating error that can't forbid illegal comibinations of QoS and priorityClass by @IULeen in #2595
• device-daemon: fix yaml by @ZiMengSheng in #2663
• community: add device & job approver by @ZiMengSheng in #2665
• koordlet: fix evict feature enabled by @lijunxin559 in #2667
• scheduler: log deviceShare allocation result by @ZiMengSheng in #2668
• webhook: support skipping pod enhanced validation via label by @TaoYang526 in #2666
• koordlet: add metric for cpuset sharepool cpus. by @tan90github in #2649
• scheduler: provide tooling for schedule diagnosis by @ZiMengSheng in #2669
• scheduler: fix reservation unreserve when binding leak by @saintube in #2673
• scheduler: optimize schedule audit with queue by @ZiMengSheng in #2676
• ci: fix libpfm4 download from sourceforge 403 by @ZiMengSheng in #2683
• scheduler: make preemption state async by @ZiMengSheng in #2682
• scheduler: revise reservation handlers and ForgetPod for multi-scheduler by @saintube in #2680
• chore: upgrade ding-talk code by @ZiMengSheng in #2687
• deps: upgrade nri to 0.7.0 by @lujinda in #2679
• scheduler: fix coscheduling prebind with illegal labels by @saintube in #2688
• scheduler: refine reservation controller args by @saintube in #2690
• e2e: bump kind version by @saintube in #2686
• scheduler: fillGPUTotalMem when designatedAllocation by @ZiMengSheng in #2695
• scheduler: no fillGPUTotalMem when mem&mem-ratio both exists by @ZiMengSheng in #2697
• scheduler: optimize deviceShare auditability for designatedAllocation by @ZiMengSheng in #2696
• scheduler & webhook: add replace-resources pod transformer and support pod transform for quota evaluation by @TaoYang526 in #2681
• scheduler: refine reservation nominator by @saintube in #2701
• koordlet: set pod priority value to default evict priority by @lijunxin559 in #2703
• webhook: support skip validating priority when feature ColocationProf… by @lijunxin559 in #2702
• chore(deps): use exported kubernetes qos package by @hunshcn in #2707
• api: fix podRequests of utils by @ZiMengSheng in #2708
• scheduler: add ReplaceResourcesTransformer feature-gate by @TaoYang526 in #2712
• scheduler: revise reservation's event handler for multi-scheduler by @saintube in #2715
• feat(gpu): report healthy condition for NVIDIA GPU devices by @summingyu in #2711
• scheduler: fix deviceshare leak for unassigned pods by @saintube in #2718
• deps: update ghw to v0.20.0 by @ZhuZhezz in #2700
• scheduler: improve dp adapter workflow by @zqzten in #2720
• koordlet: handle GPU init failure when GPU is lost by @ZhuZhezz in #2706
• koordlet: fix BECPUSuppress adjust cfs_quota failed when BE pods exist by @saintube in #2723
• scheduler: return success when no reservation reserve numa resource by @ZiMengSheng in #2724
• scheduler: revise reservation and coscheduling for multischeduler by @saintube in #2721
• scheduler: fix designated NUMA Hints by @ZiMengSheng in #2730
• koordlet: fix pod creation fails with "invalid argument"when hyperThreadEnabledRatio > 1.0 in CPU Normalization by @clay-wangzhi in #2694
• scheduler: run PreBindReservation plugins orderly by @saintube in #2733
• scheduler: fine-grained device scheduling support MetaX GPU/sGPU by @zqzten in #2734
• "k8s.io/util/pointer" is deprecated, use "k8s.io/util/ptr" instead by @ikukaku in #2684
• scheduler: fast look-up matchable reservations by @saintube in #2737
• scheduler: Introducing QueueingHint into the elasticquota plugin by @BUPT-wxq in #2717
• scheduler: fix frameworkExtender for multi-profile by @saintube in #2740
• scheduler: fix pods fail to properly use a reservation yet still consume its resources by @qinfustu in #2738
• scheduler: fix pods fail to properly use a reservation yet still consume its resources by @qinfustu in #2742
• scheduler: support designatedHints for NodeNUMAResource by @ZiMengSheng in #2745
• scheduler: refactor dp adapter node lock workflow by @zqzten in #2743
• koord-device-daemon: support for MetaX by @ZhuZhezz in #2746
• scheduler: nodenumaresource adapt pre-allocation, revise reservation AfterPreFilter by @saintube in #2744
• scheduler: ReservationCache adapt multi profile by @saintube in #2747
• Update OWNERS by @hormes in #2756
• scheduler: add questionObjectKey field for explanation by @ZiMengSheng in #2760
• slo-controller: support static reserved mode for mid resource by @lijunxin559 in #2725
• scheduler: pass node to TopologyManager interface by @ZiMengSheng in #2761
• scheduler: add ReservationNominate method by @saintube in #2757
• scheduler: deviceshare adapt pre-allocation by @saintube in #2752
• scheduler: fix filter issue for non-required pre-allocation reservation by @TaoYang526 in #2750
• apis: fix slo strategy CPUEvictThresholdPercent validate by @lijunxin559 in #2762
• koordlet: refine nri server by @saintube in #2766
• scheduler: fix representative pod stuck in unschedulablePods by @ZiMengSheng in #2773
• scheduler: revise PreBind for multischeduler by @saintube in #2774
• scheduler: nodenumaresource and deviceshare support unassigned event by @saintube in #2776
• fix(koordlet):Unbind container level cfs_quota by @AutuSnow in #2767
• scheduler: retry patch when connection close by @ZiMengSheng in #2784
• docs: add adopter & fix format by @zqzten in #2785
• scheduler: optimize when bestEffort numaTopologyPolicy by @ZiMengSheng in #2791
• add koord-queue proposal by @KunWuLuan in #2728
• scheduler: optimize failedDetail,alreadyWaitForBound for diagnosis and add ttl for explanation by @ZiMengSheng in #2792
• scheduler: fix diagnosis panic by @ZiMengSheng in #2795
• scheduler: make reservation unschedulable tolerable by @saintube in #2796
• koord-descheduler: support skip eviction gates by @songtao98 in #2770
• fix(cosSchedule):Optimize network topology Binpack effect by @AutuSnow in #2798
• koord-descheduler: fix log verbosity by @songtao98 in #2794
• Fix: collector_rdma_linux should not abort on RDMA scan errors by @PixelPixel00 in #2802
• scheduler: make network-topology aware existing pods by @ZiMengSheng in #2797
• scheduler: support configure defaultGangMatchPolicy by @ZiMengSheng in #2805
• scheduler: support designated vf by @ZiMengSheng in #2807
• proposal: Inference Orchestration Enhancement with Grove Integration by @kangclzjc in #2777
• scheduler: add nrt in topology options by @ZiMengSheng in #2808
• scheduler: export FindOneNode by @ZiMengSheng in #2814
• scheduler: support prefer node hint by @saintube in #2787
• scheduler: fix ForgetPod panic during startup by @saintube in #2827
• scheduler: fixup backup-scheduler sync issue with leader by @ZiMengSheng in #2820
• koord-descheduler: support custom priority plugin by @songtao98 in #2772
• koord-descheduler: fix AnomalyCondition check bugs and make nodePools inherit top-level default configs. by @tan90github in #2816
• chore: upgrade controller-gen from v0.14.0 to v0.20.0 by @106umao in #2819
• scheduler: revise reservation metrics by @saintube in #2829
• koordlet: support eviction trigger by allocatable by @lijunxin559 in #2780
• scheduler: audit and metric preemption duration by @ZiMengSheng in #2835
• koord-descheduler: use raw-allocatable for threshold calculation in descheduler lowNodeLoad plugin. by @tan90github in #2817
• scheduler: support cluster mode and multiple pre-allocated pods for pre-allocation by @TaoYang526 in #2768
• scheduler: make topology options RLock by @ZiMengSheng in #2841
• scheduler: refactor dryRunner for alert and explanation by @ZiMengSheng in #2842
• kubeclients: use protobuf for native resources #2812 by @manukasvi in #2815
• scheduler: fix broken reservation-ignored filtering by @saintube in #2844
• koordlet: extend nrt reporting, support numa-level reservation by @saintube in #2845
• all: Upgrade to k8s 1.35.2 by @saintube in #2830
• koordlet: refine NUMA info reporting extensions by @saintube in #2850
• scheduler: use original schedulerName for arbitrator & dryrunner by @ZiMengSheng in #2848
• descheduler: skip load-aware balancing without target nodes by @aviralgarg05 in #2838
• scheduler: log nothing when clusterNetworkTopology not found in cluster by @ZiMengSheng in #2861
• fix(OfferSlots):The number of OfferSlots should be adjusted according by @AutuSnow in #2799
• dashboard: add dashboard for scheduling and descheduling by @ZiMengSheng in #2847
• scheduler: fix flaky test in reservation plugin by @TaoYang526 in #2846
• scheduler: introduce dp adaption for huawei ascend 300i duo by @zqzten in #2789

New Contributors

• @IULeen made their first contribution in #2595
• @lujinda made their first contribution in #2679
• @hunshcn made their first contribution in #2707
• @summingyu made their first contribution in #2711
• @ikukaku made their first contribution in #2684
• @AutuSnow made their first contribution in #2767
• @PixelPixel00 made their first contribution in #2802
• @106umao made their first contribution in #2819
• @manukasvi made their first contribution in

Note truncated.

What's Changed

• koordlet: support init container for CPUSetAllocator by @zwzhang0107 in #2349
• fix(scheduler/api): register missing NodeResourcesFitPlusArgs in v1 c… by @ditingdapeng in #2353
• scheduler: optimize nodeNUMAResource performance by @ZiMengSheng in #2356
• scheduler: fix numaNodeResource not reserve cpu by @ZiMengSheng in #2355
• koordlet: skip reconcile containers in runtime hooks for kata pods by @zwzhang0107 in #2357
• scheduler: inplace update quota when quota parent change by @shaloulcy in #2351
• scheduler: update quota tree when quota spec changed by @shaloulcy in #2358
• chore: refactor ReservationOwnerMatcher func by @googs1025 in #2359
• scheduler: pre-calculate reservation info for perf by @saintube in #2366
• koordlet: BECPUSuppress supports cpuSuppressMinPercent by @saintube in #2369
• scheduler: make Activate invokable by handle by @ZiMengSheng in #2372
• chore: move SetReservationUnschedulable func to util package by @googs1025 in #2362
• feature(GCmetrics): support GCHistogramVec for util package by @googs1025 in #2382
• feat(koordlet): Add pod evict metrics by kill containers by @dongjiang1989 in #2388
• fix(koordlet):BlkIOReconcile plugin not using WaitForCacheSync method correctly to synchronize by @googs1025 in #2386
• scheduler: add log for diagnosize by @ZiMengSheng in #2397
• scheduler: add elasticquota ForgetPod, enable lazyreservationrestore for preferred topologyspread by @saintube in #2394
• chore: update actions runner image to Ubuntu 22.04 by @dongjiang1989 in #2396
• apis: add default Pending phase when creating reservation by @googs1025 in #2390
• scheduler: add metrics for elastic quota and refact UpdateQuota by @shaloulcy in #2392
• scheduler: reduce loadaware lock overhead, improve reservation args by @saintube in #2399
• scheduler: add ReservationStatusPhase metrics for reservation feature by @googs1025 in #2368
• chore: fix wrong reservation phase in unit test by @googs1025 in #2403
• webhook: add error info when failed to check sub and parent group quo… by @lijunxin559 in #2406
• scheduler: support DispenseWithLRNDeviceAllocation by @ZiMengSheng in #2410
• koordlet: add resctrl qos collector by @Rouzip in #2005
• scheduler: add OmitNodeLabelsForReservation, expose GetQuotaInformer by @saintube in #2412
• koord-descheduler: add nil check for podMetric in NodeFit of loadaware plugin by @ClanEver in #2405
• scheduler: fix reservation cache leak when terminated and deleted by @saintube in #2416
• scheduler: Add gcDurationSeconds field to ReservationArgs by @googs1025 in #2402
• webhook: check quota that only allow max >= used for specific resources by @zheng-weihao in #2409
• scheduler: change quota info to rw lock by @shaloulcy in #2421
• scheduler: rename leastRequestedScore func by @googs1025 in #2422
• scheduler: optimize secondary device well planned by @ZiMengSheng in #2425
• scheduler: avoid string concatenation and podLister in loadaware func… by @lijunxin559 in #2424
• scheduler: optimize filterNodeDevice by @ZiMengSheng in #2428
• scheduler: correct secondaryDeviceNotWellPlanned metric by @ZiMengSheng in #2431
• scheduler: remove Duplicate goroutines and remove Deprecated poll func by @googs1025 in #2429
• chore: update EnableRuntimeQuota param description by @cntigers in #2434
• scheduler: fix shared gpu binpack bug by @ZiMengSheng in #2436
• descheduler: return the unmatched cases first by @googs1025 in #2435
• chore(statesInformer): remove useless code and use generateQueryDuration() func in collectMetric() func by @googs1025 in #2439
• scheduler: support multi-scheduler by @ZiMengSheng in #2441
• scheduler: fix CheckParentQuota bug when multiquotatree by @ZiMengSheng in #2442
• e2e: disable LoadAware Filter to fix flaky test by @ZiMengSheng in #2448
• scheduler: abstract nextPod and gang inplements it by @ZiMengSheng in #2417
• chore: qos manager removes repeat goroutine by @googs1025 in #2437
• scheduler: add QuotaHookPlugin for extensible resource limiting in ElasticQuota plugin by @TaoYang526 in #2415
• scheduler: add ElasticQuota arg to control preemption of default quota (#2413) by @LennonChin in #2449
• scheduler: fix calFreeWithPreemptible modify nodeDevice by @ZiMengSheng in #2453
• scheduler: allow to disable Controllers by @saintube in #2452
• scheduler: gpuSharedPod doesn't fit secondaryDeviceWellPlanned by @ZiMengSheng in #2454
• chore: refactor ValidateNodeNUMAResourceArgs func and add miss NUMAScoringStrategy validate by @googs1025 in #2426
• chore: rename validateEstimatedResourceThresholds -> validateEstimatedScalingFactors by @googs1025 in #2427
• scheduler: fix noisy monitor timeout for unhandled dequeue pod by @saintube in #2458
• scheduler: change elastic quota Reserve/Unreserve lock by @shaloulcy in #2456
• all: enhance colocation profile by @saintube in #2445
• koordlet: support extension controllers by @shaloulcy in #2459
• scheduler: run pod-update hooks for elastic-quota regardless of whether used resources have changed by @TaoYang526 in #2461
• scheduler: add FG validatingPodDeviceResource and EnableSyncGPUShared… by @ZiMengSheng in #2466
• scheduler: Add hook plugins logic in ReplaceQuotas and OnQuotaUpdate methods of ElasticQuota plugin. by @TaoYang526 in #2465
• scheduler: make loadaware debuggable by @saintube in #2468
• scheduler: fix data race issues in GroupQuotaManager#IsQuotaUpdated and MockHookPlugin by @TaoYang526 in #2469
• manager: reconcile colocation-profile if enabled by @saintube in #2472
• proposal: heterogeneous GPU device reporting by @ZhuZhezz in #2423
• scheduler: fine-grained device scheduling support Huawei Ascend NPU (full card) by @zqzten in #2467
• koord-descheduler: support node selector for each descheduler profile by @songtao98 in #2168
• koord-descheduler: fix descheduler object limiter with multiple profiles by @songtao98 in #2200
• scheduler: fix quota webhook panic by @shaloulcy in #2473
• scheduler: fix runtime not updated when no pending pods in quota by @qinfustu in #2471
• scheduler: fix PreBind Patch for pods with same name but different uid by @saintube in #2476
• scheduler: reject schedulerName unmatched binding by @saintube in #2478
• koordlet: fix avg not being collected in AggregatedUsage by @ClanEver in #2479
• chore: fix wrong event type by @googs1025 in #2444
• scheduler: fix nodeName filter fail by @ZiMengSheng in #2482
• koordlet: fix resctrl metric releated flags's comment. by @dabaooline in #2483
• scheduler: revise PreBind safe Patch by @saintube in #2485
• scheduler: indicates X-th member pod failed when gangScheduling by @ZiMengSheng in #2486
• scheduler: support leave allocate logic to kubelet by @ZiMengSheng in #2487
• fix: GetLocalStorageInfo maybe hang by @dabaooline in #2484
• scheduler: revise event handlers for updating scheduler name by @saintube in #2488
• scheduler: put Coscheduling postFilter logic in afterPostFilter by @ZiMengSheng in #2490
• all: fix switching schedulers by @saintube in #2493
• scheduler: set PodGroup's OccupiedBy field correctly by @googs1025 in #2494
• scheduler: fix pod not update in nextpod by @ZiMengSheng in #2495
• scheduler: use RWMutex instead of Mutex by @googs1025 in #2497
• descheduler: fix podFitsAnyNodeWithThreshold when node not fit pod by @dabaooline in #2492
• scheduler: fine-grained device scheduling support Huawei Ascend vNPU by @zqzten in #2496
• scheduler: introduce gpu minors annotation to device plugin adapter by @zqzten in #2504
• manager: colocationprofile support appending label suffix by @saintube in #2503
• chores: update community meeting information by @songtao98 in #2505
• manager: fix typo for colocationprofile by @saintube in #2506
• scheduler: add koordinator.sh/gpu-memory to default device share scoring strategy resources by @zqzten in #2507
• scheduler: revise preEnqueue RepresentativePod by @ZiMengSheng in #2509
• scheduler: mark gangGroupScheduling with startTime and fix pod status by @ZiMengSheng in #2512
• koordlet: remove qos module by @Rouzip in #2440
• koordlet(statesinformer): sort podsMetricInfo and hostAppMetricInfo by name for consistent output by @googs1025 in #2510
• scheduler: fix unhandled timeout in SchedulerMonitor by @saintube in #2518
• manager: Optimize ConfigMap update event handling by reordering checks by @googs1025 in #2511
• koord-descheduler: use RWMutex instead of use Mutex by @googs1025 in #2514
• Update descheduler approvers by @songtao98 in #2520
• scheduler: add pre-allocation api, revise reservation interfaces by @saintube in #2513
• koordlet: adding cpu.max.burst comments by @bobsongplus in #2524
• colocationprofile: support namespace selector for ClusterColocationProfile in controller by @googs1025 in #2522
• webhook: add more info when deleting quota failed by @googs1025 in #2519
• scheduler: add loadaware plugins Aggregated Args Validate by @googs1025 in #2527
• proposal: network topology aware scheduling by @yccharles in #2474
• scheduler: fix flaky test by @ZiMengSheng in #2532
• scheduler: loadaware support force estimated duration and other improvements by @zheng-weihao in #2531
• remove inactive approve to smooth review process by @ZiMengSheng in #2534
• koordlet: fix Histogram maxValue error by @cntigers in #2540
• scheduler: add ElasticQuota arg to control MinQuotaScale by @qinfustu in #2542
• scheduler: only requeue gang when gangWorth and by activate by @ZiMengSheng in #2537
• koord-manager: support enhanced validation for pod by @TaoYang526 in #2462
• scheduler: fix missing Unreserve when ResizePod=true by @saintube in

Note truncated.

What's Changed

• fix(scheduler/api): register missing NodeResourcesFitPlusArgs in v1 c… by @ditingdapeng in #2353
• scheduler: fix the noderesourcefitplus plugin does not implement the LocalStorageCapacityIsolation FeatureGate by @qinfustu in #2536

Full Changelog: v1.6.0...v1.6.1

What's Changed

• scheduler: refine reservation msg if matched filtered out by other plugins by @saintube in #2085
• Add ZiMengSheng into approvers alias by @ZiMengSheng in #2087
• scheduler: fix reservation plugin clone concurrent read write by @buptcozy in #2084
• scheduler: fix elastic quota TestPlugin_OnPodDelete random panic by @buptcozy in #2083
• scheduler: change reservation event message format by @zwzhang0107 in #2090
• scheduler: add node info in reservation level event by @zwzhang0107 in #2094
• scheduler: support consume reserved numa resource by @ZiMengSheng in #2080
• scheduler: add new pod estimate with loadaware plugin by @zwForrest in #1992
• koord-descheduler: limit the total number of pod evictions by @zwForrest in #2091
• chore: fix e2e configs after the load-aware scheduling updated by @saintube in #2101
• koord-descheduler: balance prod pods between nodes in LowNodeLoad by @zwForrest in #2066
• chore: fix low node load test after prod thresholds added by @songtao98 in #2103
• koord-scheduler: Pod updates should not update the timestamp by @zwForrest in #2100
• koord-descheduler: fix object limiter by move it to reconciler by @songtao98 in #2088
• koord-descheduler: enhance LowNodeLoad scorer by @LY-today in #2092
• koord-descheduler: support evicting all bare pods in migration controller by @songtao98 in #2102
• koordlet: update resctrl qos proposal by @kangclzjc in #2079
• koordlet: add taskids in statesinformer by @kangclzjc in #2057
• koord-manager: fix the prom metrics handler by @saintube in #2107
• koord-descheduler: fix descheduler log by @zwForrest in #2114
• scheduler: support pod request exact match reservation by @ZiMengSheng in #2121
• scheduler: reduce cycleState overhead for reservation by @saintube in #2120
• scheduler: fix reserveNUMAResource bug by @ZiMengSheng in #2124
• koordlet: add allpods reconciler func by @kangclzjc in #2112
• scheduler: reduce deviceShare memory overhead by @ZiMengSheng in #2126
• webhook: add quota evaluate webhook by @shaloulcy in #2129
• koordlet: upgrade nri to 0.6.1 by @kangclzjc in #2132
• koord-manager: support manually configured default cpu normalization settings by @yangfeiyu20102011 in #2128
• koordlet: tc plugin for netqos by @lucming in #1920
• chore: fix koordlet dockerfile by @saintube in #2134
• koord-descheduler: Fix threshold initialization for prod. by @zwForrest in #2130
• scheduler: support device topology strategy by @ZiMengSheng in #2133
• scheduler: optimize when only 1 reservation/node by @ZiMengSheng in #2136
• koord-descheduler: add migration object limiter for namespace by @songtao98 in #2068
• webhook: revise elasticquota mutating when validating disabled by @saintube in #2135
• koord-descheduler: support max migrating globally limitation by @songtao98 in #2143
• scheduler: skip uncessary Filter&Score plugin by @ZiMengSheng in #2138
• all: support priority and preemption policy transformer by @saintube in #2137
• scheduler: elastic quota ignore terminating pod by @shaloulcy in #2141
• koord-descheduler: fix maxMigratingGlobally arg json tag by @songtao98 in #2144
• scheduler:support multi gpu share by @AdrianMachao in #2127
• scheduler: return framework.Unschedulable when cpuPolicy not satisfied by @ZiMengSheng in #2146
• koord-manager: fix invalid assignments in the resource amplification … by @yangfeiyu20102011 in #2149
• koordlet: add some information to improve log readability by @yangfeiyu20102011 in #2153
• ci: add disk GC for E2E jobs by @saintube in #2157
• scheduler: fix panic when NUMANode equals -1 by @ZiMengSheng in #2151
• scheduler: fix too many ut log by @ZiMengSheng in #2154
• scheduler: support scheduler config v1 by @AdrianMachao in #2167
• scheduler: add coscheduling preEnqueue by @AdrianMachao in #2155
• koordlet: add record events by @kangclzjc in #2162
• koordlet: fix unsafe conversion in net_cls by @saintube in #2173
• scheduler: assure only gpu will allocate by topology by @ZiMengSheng in #2178
• schedler: move gang OnceSatified to gangGroupInfo by @buptcozy in #2176
• scheduler: add reservation preemption by @saintube in #2139
• koord-descheduler: update k8s descheduler to 0.28.0 by @songtao98 in #2156
• scheduler: support reservation ignored and nodenumaresource preemption by @saintube in #2163
• koord-manager: support resource amplification config, cpu, memory and other resource by @yangfeiyu20102011 in #2172
• scheduler: fix pod update when pod creation is before quota creation by @shaloulcy in #2177
• scheduler: permit other plugin to set numaAffinity by @ZiMengSheng in #2182
• koord-descheduler: LowNodeLoad check if evicted pod can cause new node over utilized by @songtao98 in #2142
• Update OWNERS_ALIASES by @hormes in #2145
• koord-descheduler: fixes namespace object limiter by @songtao98 in #2160
• scheduler: suppport amd.com/gpu by @ZiMengSheng in #2174
• scheduler: elastic quota ignore terminating pod immediately by @TaoYang526 in #2180
• apis: add reservation name, taints and tolerations in ReservationAffinitty by @zwzhang0107 in #2186
• scheduler: fix panic when nonimatingInfo nil by @ZiMengSheng in #2189
• apis: add protobuf for reservation by @zwzhang0107 in #2192
• scheduler: fix ElasticQuota state's Clone by @saintube in #2190
• koordlet: fix wrong msg in calculateBESuppressCPU by @yangfeiyu20102011 in #2193
• chores(deps): Add support depguard rules in golangci-lint by @dongjiang1989 in #1965
• koord-descheduler: fix migration controller max unavailable computing algorigthm by @songtao98 in #2196
• scheduler: support devices of the same node gpuMem not equal by @ZiMengSheng in #2199
• scheduler: support pod preemption from numa awareless reservation by @ZiMengSheng in #2204
• scheduler: optimize numa affinity store by @ZiMengSheng in #2209
• apis: fix type of MinResources in podgroup by @zwzhang0107 in #2212
• koordlet: Add resctrl runtime hook for pod level by @kangclzjc in #2123
• koord-manager: nodeslo-controller enqueue request when node labels updated by @chengjoey in #2201
• scheduler: optimize performance on transformer extension and Skip status by @saintube in #2211
• koord-manager: add unallocated resource into mid resource. by @tan90github in #2152
• scheduler: support reservation name, taints and tolerations by @saintube in #2207
• scheduler: recover gang check in preFilter by @ZiMengSheng in #2217
• apis: GPU Partition related by @ZiMengSheng in #2219
• scheduler: optimize GPU allocate logic by @ZiMengSheng in #2221
• scheduler: support amd by @ZiMengSheng in #2223
• scheduler: support reserving pods resources by @saintube in #2224
• scheduler: allocate tolerate numa-meaning-less device by @ZiMengSheng in #2226
• scheduler: fix partition binpack disorder by @ZiMengSheng in #2227
• koord-descheduler: fix prod pod excessive eviction when node recover to normal by @JBinin in #2225
• scheduler: fix gpu shared bug by @ZiMengSheng in #2228
• scheduler: support secondary device well planned by @ZiMengSheng in #2229
• scheduler: fix shared gpu pod allocated minor of -1 by @ZiMengSheng in #2230
• scheduler: improve gang log by @googs1025 in #2009
• scheduler: consider pod requests when gpu&RDMA joint allocate by @ZiMengSheng in #2233
• scheduler: optimize reservation perf with lazy restoring by @saintube in #2241
• util: fix incorrect comparison for resource.go#LessThanOrEqualCompletely by @TaoYang526 in #2235
• scheduler: support besteffort policy by @ZiMengSheng in #2243
• scheduler: fix scaling factor 100 for burstable pod by @ZiMengSheng in #2242
• gpu: setting default gpu partition policy by @ZiMengSheng in #2245
• koordlet: add ReadMemoryUsage method to improve generality of cgroup reader interface by @yangfeiyu20102011 in #2251
• koord-manager: refactor oversale resource calculate logic. by @tan90github in #2240
• util: add defaulting to blkio qos to improve robustness by @zqzten in #2238
• webhook: Support setting quota admission to zero for special usage scenarios, such as temporarily pausing pod submissions. by @TaoYang526 in #2237
• scheduler: add downgrade strategy for empty 'aggregated' on cold koor… by @clay-wangzhi in #2239
• koord-manager: mv slo-controller metrics to util by @zwzhang0107 in #2257
• scheduler: fix deviceshare with reservation-ignored pods by @saintube in #2252
• scheduler: setFailedPlugin when plugin transformer failed by @ZiMengSheng in #2262
• koordlet: export host application cpu and memory usage for prometheus by @yangfeiyu20102011 in #2259
• scheduler: update dingtalk QR Code by @ZiMengSheng in #2277
• scheduler: support allocating from reservation when no resource matched by @saintube in #2279
• koordlet: add GetNodeMetricSpec in statesInformer interface by @j4ckstraw in #2278
• api: supply id for minor meaningless device by @ferris-cx in #2250
• koordlet: remove useless stateInformer.impl StateInformer interface by @j4ckstraw in #2286
• Upgrade ginkgo to v2.11.0 by @nce3xin in #2284
• koordlet: support change pod CPUQOS by annotations by @j4ckstraw in #2206
• koordlet: rdma device inject by @ferris-cx in #2285
• koord-manager: consider NodeReserved when calculate mid resource. by @tan90github in

Note truncated.

What's Changed

• koordlet: fix core sched conflicts with GI and revise API by @saintube in #1829
• koordlet: fix recursively disabling bvt by @saintube in #1848
• koordlet: fix struct type lost of node slo extension by @zwzhang0107 in #1847
• scheduler: fix ElasticQuota creation failure on startup by @eahydra in #1845
• koordlet: eliminate cache GC is not started by @saintube in #1849
• scheduler : supports disable runtime quota by @xulinfei1996 in #1839
• chore(deps): bump actions/cache from 3 to 4 by @dependabot in #1851
• scheduler: optimize ElasticQuota status/metrics sync flow by @eahydra in #1850
• scheduler: fix ElasaticQuota sync status missing decorate used by @jasonliu747 in #1852
• scheduler: fix missing tree id in GetQuotaSummary() by @jasonliu747 in #1853
• scheduler: ElasticQuota runtime is no longer calculated when not needed by @eahydra in #1855
• koord-manager: refactor gpu device resource plugin by @saintube in #1846
• descheduler: fix bug of nodeAnomalyDetector in LowNodeLoad plugin by @bogo-y in #1858
• scheduler: improve plugin args by @eahydra in #1857
• scheduler: improve Coscheduling PreFilter status by @eahydra in #1861
• Add a new adopter by @zhifanggao in #1865
• metrics: seperater metrics as internal and external for slo-controller and koordlet by @zwzhang0107 in #1807
• koordlet: fix metric record err by @ls-2018 in #1868
• scheduler: fix the method of calculating desiredCount in DeviceShare ApplyForAll strategy by @eahydra in #1870
• webhook: add node affinity for pod when MultiQuotaTree on by @shaloulcy in #1864
• webhook: validate network qos config in webhook by @lucming in #1841
• proposal: Resctrl Qos Enhancement Proposal by @kangclzjc in #1798
• koordlet: decouple CUDA dependency Accelerators feature disabled by @saintube in #1876
• koordlet: export mid resource metrics by @j4ckstraw in #1874
• koordlet: revise base collectors and system status check by @saintube in #1877
• koordlet: fix proxy container service and revise cpu.max updater by @saintube in #1885
• webhook: improve node affinity performance for MultiQuotaTree by @shaloulcy in #1872
• scheduler: correct NodeNUMAResource handling with NodeFullPCPUsOnly and preferredCPUBindPolicy by @eahydra in #1891
• koordlet: fix misleading query helper by @saintube in #1893
• chore(deps): bump helm/kind-action from 1.8.0 to 1.9.0 by @dependabot in #1897
• webhook: more checks for netqos by @lucming in #1878
• koorldet: remove logging field from the kubelet configz by @zwzhang0107 in #1907
• proposal: support Job-level preemption by @xulinfei1996 in #1879
• proposal: support Reservation preemption by @xulinfei1996 in #1887
• webhook: fix webhook for root quota creation by @shaloulcy in #1916
• scheduler: fix initializing built-in quota objects by @eahydra in #1917
• Proposal: add proposal for multi tree quota by @shaloulcy in #1902
• webhook: fix validateQuotaTopology for root quota by @shaloulcy in #1918
• webhook: forbid use internal protocols by @ls-2018 in #1911
• webhook: remove some annotation verify by @ls-2018 in #1921
• scheduler: add Reservation restricted options to control allocatable resources by @eahydra in #1894
• koordlet: add metrics for kubelet and resource executor by @zwzhang0107 in #1913
• scheduler: remove the restriction that reservation cannot be preempted by @eahydra in #1859
• scheduler: fix reservation restricted options by @eahydra in #1926
• scheduler: add unschedulable resource for root quota by @shaloulcy in #1925
• scheduler: improve NodeNUMAResource handling node cpu bind policy by @eahydra in #1892
• utils: add metric server client by @zwzhang0107 in #1881
• scheduler: update resourceNames of ReservationInfo by @eahydra in #1930
• koordlet: support NetworkQoS plugin by @l1b0k in #1843
• koordlet: revise cgroupRoot param in podsInformer by @saintube in #1928
• koordlet: optimize perf group cpi collection by @bowen-intel in #1905
• koordlet: support blkio-cost rpct and wpct configuration by @wangxiaoq in #1931
• scheduler: resolve the wrong use of gangmod in gang init phase by @PeterChg in #1934
• Add a new adopter by @j4ckstraw in #1938
• scheduler: revise device allocate logic by @ZiMengSheng in #1927
• webhook: fix concurrent map for quota topology by @shaloulcy in #1940
• koordlet: fix bug that cpu eviction does not take effect by @wangxiaoq in #1932
• scheduler: extend reservation nominator to support reservation preemption by @xulinfei1996 in #1936
• koordlet: fix GetAllPods error and wrong tsdb prometheus registry by @saintube in #1944
• proposal: support pod customizing numa policy by @KunWuLuan in #1910
• manager: add featuregate to remove the restriction of parent quota submit pods by @xulinfei1996 in #1929
• scheduler: improve coscheduling's PostFilter if not collect enough by @eahydra in #1948
• scheduler: fix reservation calc error by @eahydra in #1949
• koordlet: support blkio-cost user cost model configuration by @wangxiaoq in #1942
• scheduler: improve the failed event of Reservation's scheduling by @ZiMengSheng in #1947
• scheduler: pick vf by random by @ZiMengSheng in #1953
• scheduler: give more context to device allocator by @ZiMengSheng in #1958
• descheduler: change the order of check len(lowNodes) and len(sourceNodes) by @bogo-y in #1957
• descheduler: add a cache timeout argument for nodeAnomalyDetectors in lowNodeLoad plugin. by @bogo-y in #1955
• descheduler: only include the overused resources when sorting removable pods by @bogo-y in #1962
• apis: add Recommendation crd by @zwzhang0107 in #1937
• webhook: fix priority admission conflict by @j4ckstraw in #1968
• apis: add Resctrl API by @kangclzjc in #1943
• chore: fix make docker build xxx by @j4ckstraw in #1950
• scheduler : fix deviceshare plugin of add\remove pod by @xulinfei1996 in #1963
• Revert "scheduler : fix deviceshare plugin of add\remove pod (#1963)" by @xulinfei1996 in #1971
• koordlet: record invoke metrics for runtime hook by @saintube in #1961
• scheduler: optimize QueueSort func when Gang and Barepod Coexists by @ZiMengSheng in #1977
• scheduler: fix reservation nominator residual bug by @xulinfei1996 in #1985
• Add codecov token by @FillZpp in #1987
• slo-controller: refactor codes for reading by @hormes in #1973
• CI workflow with codecov token by @FillZpp in #1989
• chores: fix some fmt err by @ls-2018 in #1970
• Revert "scheduler: pick vf by random (#1953)" by @ZiMengSheng in #1986
• proposal: tc plugin for netqos by @lucming in #1976
• scheduler: assure Fairness and DeadlockFree by @ZiMengSheng in #1996
• scheduler: sort gang of same gangGroup by gangId by @ZiMengSheng in #1997
• apis & slo-controller: allow specifying node-wise total bandwidth via annotation by @sjtufl in #1982
• scheduler: no rejecting sbiling when invalid scheduling cycle by @ZiMengSheng in #1999
• descheduler: allow annotated pod pass non-retrievable filter by @ZiMengSheng in #1994
• scheduler: add more quota details used and request in quotaInfo by @xulinfei1996 in #2001
• chores: add security docs by @zwzhang0107 in #2007
• koordlet: add close for querier by @zwzhang0107 in #1995
• koordlet: add extended HTTP handler registry by @saintube in #2006
• chore: fix gpu api in comments by @hormes in #1988
• scheduler: quota info deepcopy withou pod deepcopy by @xulinfei1996 in #2008
• scheduler: refine coscheduling fairness consider gangGroup by @buptcozy in #2004
• scheduler: optimize scheduleCycle logic when notEnoughChildren by @ZiMengSheng in #2011
• scheduler: fix panic of podGroup and pod delete order issue by @buptcozy in #2012
• fix panic of podGroup and pod delete order issue2 by @buptcozy in #2013
• scheduler: refine restricted topology policy by @ZiMengSheng in #2010
• koordlet: revise node topology reporting for the kubelet cpu manager by @saintube in #2015
• koordlet: support cri-o container runtime by @georgexiang in #1983
• scheduler: try best to distribute cpu and memory evenly across numa by @ZiMengSheng in #2017
• scheduler: remove invalid hint in which some numaNode lack resource by @ZiMengSheng in #2018
• chores: add openssf best practices badge into README by @songtao98 in #2023
• webhook: optimize webhook patchResponse function by @ZiMengSheng in #2025
• utils: change metrics server util pkg name by @zwzhang0107 in #2016
• descheduler: Add detail logs for nonRemovable pods by @zwForrest in #2000
• chore: fix some nit error by @googs1025 in #2027
• chore(deps): bump golangci/golangci-lint-action from 3 to 5 by @dependabot in #2022
• scheduler: coscheduling plugin only record gang OnceResourceSatisfied… by @xulinfei1996 in #2029
• scheduler: fix that cpu should be preferred if numa policy is restricted by @KunWuLuan in #2033
• koordlet: add NRI reconnect by @kangclzjc in #1972
• scheduler: support numa topology policy on pod by @KunWuLuan in #1939
• all: migrate to 1.28.7 by

Note truncated.