koordinator-sh / koordinator

A QoS-based scheduling system brings optimal layout and status to workloads such as microservices, web services, big data jobs, AI jobs, etc.

https://koordinator.sh

koordinator-sh/koordinator Issues

[install error] koordlet panic
Updated 4 days ago3
[BUG] v1beta3 doesn't support PreEnqueue extension point !!
Updated 6 days ago1
[proposal] cpu burst should also burst some pods when the node util nears threshold
Updated 7 days ago
[proposal] Coscheduling move some BeforePreFilter quick check into PreEnqueueExtension
Updated 7 days ago5
Continously getting "no space left on Device" even though node has enough space.
Updated 7 days ago2
[proposal] How to Better Integrate Mid-Tier Runtime Hooks with Batch Resources?
Updated 8 days ago1
[BUG] NodeNUMAResource should support allocating reserved cpus when CPU amplification enabled
Updated 9 days ago
[BUG] Pods been deleted from gang.Children when recreated with same name.
Updated 10 days ago2
[proposal] Reservation support binding to scheduled pods
Updated 14 days ago
[BUG]koord-manager failed to renew lease
Closed 15 days ago6
[proposal] runtime hook should support initContainers
Updated 15 days ago
[BUG] Koordinator doesn't support multiple card sharing
Closed 16 days ago8
[BUG] 添加了 Reservation Affinity 的 Pod 通过了调度器 Filte 阶段，但是节点上实际没有 Reservation 可以满足资源条件
Updated 16 days ago
[BUG] EnqueueRequestForNode should not only watche update of node's allocatable
Updated 16 days ago4
[proposal] koord-descheduler: Optimize the reassessment of single-node resource left
Updated 16 days ago1
[proposal] koord-descheduler support objectLimiter for namespace
Closed 24 days ago2
elastic quota 弹性抢占疑问[question]
Closed a month ago4
[question] resource reservation does not consider scale up via cluster-autoscaler
Updated a month ago4
[proposal] Report node allocatable RDT ctrl groups
Updated a month ago1
[dependency] Upgrade github.com/containerd/nri to v0.6.1
Closed a month ago2
[BUG] The parent queue has over the maximum
Updated a month ago1
[BUG] unexpected CPU eviction
Closed 2 months ago2
[question]koordinator yarn跑实时流计算作业
Closed 2 months ago2
[BUG] hadoop-yarn rm get /proxy/application_xxx java.net.BindException: Cannot assign requested address (Bind failed)
Closed 2 months ago2
[BUG] koord-manager metrics externtal exporter error
Closed 2 months ago
Performance Collector Scope and specific OS Requirements
Updated 2 months ago2
[BUG] 添加了 Reservation Affinity 的 Pod 通过了调度器 Filte 阶段，但是节点上实际没有 Reservation 可以满足资源条件
Closed 2 months ago2
[question]reservation对象和namespace配额疑问
Updated 2 months ago1
[question] elastic quota allow-lent-resource 没有生效
Updated 2 months ago1
[proposal] koord-descheduler: add a new arg evictAllBarePods to MigrationController
Closed 2 months ago1
[BUG] koord-descheduler: objectLimiter.workload doesn't work
Closed 2 months ago1
[proposal] deschedule production pods between nodes
Closed 2 months ago5
[proposal] koordinator-descheduler has ability to limit the total number of pod evictions
Closed 2 months ago2
Device topology awareness and co-scheduling of GPU and RDMA
Updated 2 months ago1
[BUG] yarn-copilot resource update failed on unrecognized node: hadoop-yarn-nm-2
Closed 2 months ago3
[BUG]koord-manager failed to renew lease
Closed 2 months ago1
[proposal] Optimize NUMA Topology Alighment Mechanism
Closed 2 months ago
[BUG] Elastic Quota Management not working as expected
Updated 2 months ago2
plugin GroupIdentity is not supported by system
Closed 2 months ago2
[proposal] CPUNormalization can amplify cpu via cfs_quota
Updated 3 months ago
[BUG] Pod Cannot Start When CPUNormalization enable
Closed 3 months ago6
REQUEST: New membership for <lucming>
Closed 3 months ago4
REQUEST: New membership for <Suozz>
Closed 3 months ago4
[PodGroup update phase Error] PodGroup cannot Update phase from Scheduling to Scheduled when gangMember larger than two
Closed 3 months ago4
[question] deploy failed to try resolving symlinks in path "/var/log/pods/koordinator-system_koord-manager-/manager/200.log": lstat : no such file or directory
Closed 3 months ago2
[No space left on device Error] koordlet is throwing No space left
Closed 3 months ago2
[proposal] Support multi-queue mechanism, similar to what Volcano does.
Updated 3 months ago
refine coscheduling fairness consider gangGrouop and scheduleTime
Closed 3 months ago2
MultiArch Support
Closed 4 months ago4
NRI Configuration status check
Closed 4 months ago4