HAMi Enterprise 2.11.0-r0
| Chart 版本 | Chart 声明的 appVersion | 发布日期 |
|---|---|---|
2.11.0-r0 | 2.11.0-r0 | 2026 年 11 月 12 日 |
本版本以开源 HAMi v2.11.0 为基线。企业版升级变化以 v2.10.0-r5 为比较基线。
软件变更
本版重点:Ascend 企业级共享能力增强
HAMi Enterprise 将 NVIDIA 上已有的企业级资源控制能力扩展到 Ascend,在 NPU 上提供相近的动态显存管理、Turbo 模式和任务优先级控制。调度器负责分配与审批,Ascend 设备插件和 libvnpu 在工作负载运行时执行控制。
- 动态显存:使用 hami-core 软切分的业务容器可通过
NPU_MEMORY_AUTO_ADJUST=true申请运行时显存扩容。调度器检查设备可用显存及命名空间配额,批准后由插件更新运行时额度。当前仅支持每个业务容器一个 vNPU,只扩不缩,不改变 Pod 的初始资源声明;拒绝或超时后保留原额度,原分配请求仍可能返回 OOM。 - Turbo 模式:
NPU_TURBO_MODE=true允许可信、算力独占的任务不受运行时 AI Core 节流,保留显存限制、记账和经审批的显存扩容。该模式默认关闭,任务应申请 100% AI Core,并通过重建 Pod 切换;Turbo 任务不参与任务优先级仲裁。 - 任务优先级控制:在
NPU_CORE_UTILIZATION_POLICY=force下,通过NPU_TASK_PRIORITY配置共享 NPU 任务的优先级。无符号整数越小优先级越高,0高于默认值1。优先级与算力份额分别配置;运行库在工作提交边界协作式协调高低优先级任务,已提交的计算自然完成,不保证固定抢占延迟。
本版配套采用 Ascend Device Plugin v1.5.0(Chart 0.4.0)和 libvnpu v1.5.0。它们独立部署,升级主 Chart 不会自动更新 Ascend 插件或已有业务容器中的运行库。动态扩容还要求显式使用 huawei.com/vnpu-mode: hami-core、Kubernetes 1.30 及以上,并启用扩容审批和准入保护。配套升级与运行库状态迁移见 Ascend Chart 0.4.0 发布说明。各项能力相对企业版 r5 的变化见下文对照表。
相对企业版 v2.10.0-r5 的变化
企业版改进与修复
- 新增 NVIDIA Downward 运行时资源边界校验:在 Kubernetes 1.30 及以上、启用准入 webhook 时,检查容器运行时设备限制是否超出其资源请求。使用自定义准入清单的部署需接入
/downward/validate。 - 修复 NVIDIA 普通初始化容器的准入处理,不再为其注入 Kubernetes 不允许的
postStart生命周期钩子;容器已有 HTTP、TCP 或 Sleep 类型的postStart时,也不再混入 Exec 动作。 - 修复 NVIDIA 显存自动扩容在包含初始化容器的 Pod 中定位业务容器分配记录的问题,避免扩容反馈更新错误的容器记录。自动扩容仍需显式启用,初始化容器不在该能力范围内。
- 更新配套 NVIDIA 运行库,修复并发显存预留与失败回滚,以及异步分配、CUDA Graph 和虚拟内存管理(VMM)相关的显存计账,并改进 CUDA/NVML 接口兼容性。
开源同步带来的变化
- 新增原生边车容器(native sidecar)的设备资源计账,将持续运行的边车容器与业务容器共同计入显存、算力及设备槽位占用。初始化阶段的峰值、初始化结束后的占用和调度器重启后的资源恢复按容器生命周期处理。
- 改进动态 MIG 的实例放置、空闲实例复用和回收,增加 MIG profile 偏好及动态 MIG 的 CDI 注入。新增 MIG 显存和利用率采集,需设置
devicePlugin.monitor.migInstanceAccess: true;CDI 注入需配套支持 CDI 的容器运行时。 - 新增可选的 NVIDIA NUMA 分配校正。当 kubelet 的拓扑约束与原分配不一致时,由设备插件请求调度器重新计算并更新分配及资源计账,默认关闭。启用时需配套设备拓扑、节点首选分配、Pod NUMA 策略和调度器 CA,参数见本页 Values 说明。
- 扩展 AMD 调度,支持显存百分比、分配阶段的命名空间显存配额校验、XGMI 与 NUMA 拓扑选择,以及可配置的 RuntimeClass。
- 新增基于 Lupine 的远程 GPU 调度,可按节点标签选择服务池,默认关闭。远程 GPU 服务端及客户端运行环境需单独配置。
- 新增 Ascend950DT 9582 的整卡调度配置,默认资源名为
huawei.com/Ascend950DT,每卡配置 98,304 MiB 显存和 32 个 AI Core。需配套独立部署的兼容设备插件;该配置不代表其他 950DT SKU 或切分模式均已支持。 - 新增可选的 Pod 首选节点亲和性评分,以及调度器 leader、缓存同步状态和 GPU 硬件健康指标;调整监控指标的弃用标记,迁移方式见下文。
- 改进配额显存单位换算、节点锁恢复、初始化容器的设备锁定,以及 Pod 删除和调度器重启时的资源记账。
相对开源 HAMi v2.11.0 的企业版扩展
| 能力 | 相对开源 v2.11.0 的增加 | 相对企业版 v2.10.0-r5 的变化 | 交付组件与使用条件 |
|---|---|---|---|
| Ascend hami-core 运行时显存扩容 | 在静态软切分基础上,增加运行时扩容申请、设备显存与命名空间配额检查,以及批准后额度回写。 | 延续已有能力;适配本版设备计量与初始化容器计账。 | 主软件及独立交付的 Ascend 插件、libvnpu;要求 huawei.com/vnpu-mode: hami-core、NPU_MEMORY_AUTO_ADJUST=true,每个业务容器仅分配一张 vNPU,并满足 Kubernetes 1.30 及以上的准入条件。 |
| Ascend Turbo 模式 | 为可信、算力独占任务提供保留显存控制的非节流运行模式。 | 延续配套运行库已有能力。 | Ascend 插件与 libvnpu;显式设置 NPU_TURBO_MODE=true,申请 100% AI Core;默认关闭,不参与任务优先级仲裁,切换需重建 Pod。 |
| Ascend 任务优先级控制 | 共享 NPU 任务可按优先级协调工作提交,算力份额与任务优先级独立配置。 | 延续配套运行库已有能力。 | libvnpu 的 force 策略与 NPU_TASK_PRIORITY;小值优先,默认 1。在提交边界协作式协调,不保证固定抢占延迟。 |
| Ascend hami-core 算力策略与容量配置 | 提供 spatial、dynamic、fixed、overdraft 策略及算力容量倍率配置,分别控制运行时算力使用方式与注册容量。 | 延续 r5 已提供的策略和配置接口。 | 主 Chart 生成共享配置,独立 Ascend 插件和 libvnpu 执行;devices.ascend.hamiCorePolicy 默认 spatial,devices.ascend.deviceCoreScaling 默认 1。容量倍率不改变单个工作负载每卡最多申请 100% 算力的限制;spatial 不代表独占硬件分区。 |
| NVIDIA 容器级分配接口 | 提供 Downward API 分配路径,通过容器级标注传递设备、显存和算力配置,供运行时及平台集成使用。 | 分配接口延续;新增 /downward/validate 运行时资源边界校验。 | 主软件;scheduler.useDownward: true 为默认值。新增校验要求 Kubernetes 1.30 及以上且启用准入 webhook。 |
| NVIDIA 运行时显存超分 | 配套运行库提供显存超分能力,可配合调度侧的显存容量配置使用。单纯提高调度器可分配容量不等于启用运行时超分。 | 延续已有能力;本版修复相关运行库的显存预留、回滚及计账。 | 企业版 NVIDIA 运行库与设备分配组件;适用的 CUDA 分配路径和工作负载需与超分配置匹配。 |
| 超分场景的 CUDA IPC 兼容处理 | 提供可选的 CUDA IPC 兼容策略,减少超分模式对部分 IPC 分配的影响。 | 延续已有能力,不作为本版新增。 | 企业版 NVIDIA 运行库;策略默认关闭,与超分配置配套使用,不代表所有 IPC 工作负载均兼容。 |
| NVIDIA 运行时显存自动扩容 | 根据运行时反馈增加业务容器的显存额度,由调度器按可分配显存检查请求并更新分配记录。 | 延续已有能力;修复 Pod 含初始化容器时的业务容器记录定位。 | 主软件与企业版 NVIDIA 运行库;业务容器显式设置 CUDA_MEMORY_AUTO_ADJUST=true,额度按 1 GiB 向上取整,不包含初始化容器。 |
| 平头哥 T-Head PPU 共享调度 | 增加 PPU 的显存、算力共享调度,以及型号、UUID、拓扑选择、配额与授权接入。 | 延续已有能力,不作为本版首次接入。 | 主软件中的 PPU 后端与独立 PPU 插件;需保持两端资源名及切分配置一致,主 Chart 不安装 PPU 插件。 |
| 企业授权管理 | 支持按设备实例或设备型号数量授权,提供授权状态、有效期和异常事件。 | 延续 r5 已有的多厂商授权能力。 | 主软件与配套运行组件;使用覆盖目标集群、设备范围和有效期的企业版许可证。 |
| 企业版资源与显存观测 | 增加显存超分倍数、算力分配方式、扩容启用及完成状态,以及上下文、模块、缓冲区显存明细。 | 延续企业版指标;本次同步的硬件健康与采集状态指标另见下文。 | scheduler、monitor 与配套运行库;扩展包括 hami_gpu_memory_oversell_factor、分配指标的 compute_allocation_mode、vgpu_mode、oom_expand、oom_expanded 标签,以及 hami_vgpu_memory_context_bytes、hami_vgpu_memory_module_bytes、hami_vgpu_memory_buffer_bytes。 |
已同步的主要开源改动
| 改动 | 对使用或集成的影响 | 对应开源 PR |
|---|---|---|
| 原生边车容器与初始化容器计账 | 计入持续运行的边车容器,并对仅初始化容器申请设备的 Pod 执行设备锁定。 | #2723、#2755 |
| MIG 放置与 profile 偏好 | 改进多实例放置;Pod 可用 nvidia.com/mig-profile-preference 指定允许列表中的 profile 偏好,不改变显存需求和硬件约束。 | #2855、#3014 |
| 动态 MIG 生命周期及 CDI | 复用匹配的空闲实例,按新分配需要回收阻挡位置的空闲实例,并维护动态 MIG 的 CDI 条目。 | #3128、#3070 |
| MIG 监控 | 通过 NVML/DCGM 采集 MIG 工作负载指标,需启用 monitor 的 MIG 设备访问权限。 | #3270 |
| NUMA 分配校正 | 增加可选的分配重算与资源预留更新;请求使用设备插件 ServiceAccount 身份认证。 | #2731、#2882 |
| AMD 显存与配额 | 支持显存百分比,并按实际分配显存检查命名空间配额。 | #3179、#3180 |
| AMD 拓扑与 RuntimeClass | 按 XGMI、NUMA 关系选择设备;仅在 Pod 未设置 RuntimeClass 时使用配置的默认值。 | #3191、#3194 |
| 远程 GPU | 增加 Lupine 服务发现、分配及服务池选择;一次分配预留同一服务器的 GPU 资源。 | #3004、#3013、#3195 |
| Ascend950DT 9582 | 增加该 SKU 的整卡配置,设备插件独立部署。 | #3316 |
| 首选节点亲和性评分 | 可通过权重使 Pod 的首选节点亲和性参与 HAMi 评分,默认关闭。 | #3298 |
| 调度状态与 GPU 健康指标 | 增加 leader、缓存同步、温度、功耗和 ECC 错误指标。 | #2758、#2732 |
| 监控指标弃用 | 保留现有输出,提供显存指标与配额上限标签的迁移方向。 | #3273、#3299 |
| Helm 厂商配置统一 | 厂商参数移至 devices.<vendor>,默认设备配置由 Values 生成。 | #3124、#3152 |
监控指标变化
| 指标或标签 | 本版状态与操作 |
|---|---|
hami_container_device_memory_bytes | 已弃用但仍输出。看板与告警改用 hami_vgpu_memory_used_bytes,标签保持一致。 |
hami_resource_quota_used 的 limit 标签 | 已弃用但仍保留。配额上限改读 hami_resource_quota_limit,按 namespace、quota_name 关联;使用 sum without (limit) 去除 usage 序列的旧标签后再做比较。 |
hami_scheduler_is_leader、hami_scheduler_cache_synced | 新增状态指标,1 表示当前为 leader 或缓存已同步,0 表示否。 |
hami_vgpumonitor_collect_success | 新增采集状态指标。0 表示采集存在失败,不能把缺失样本当成零使用量;不支持的可选硬件指标会跳过。 |
hami_host_gpu_temperature_celsius、hami_host_gpu_power_usage_watts、hami_host_gpu_ecc_errors_total、hami_host_gpu_memory_controller_utilization_ratio | 新增硬件指标,按设备及驱动支持情况采集。 |
Chart 变更
- 统一厂商参数到
devices.<vendor>,旧字段及新旧混用均在渲染时报错。设备配置中的 NVIDIA 默认资源量、MIG profile 及厂商型号列表可通过对应 Values 设置。 - 主组件默认镜像标签更新为
v2.11.0-r0。镜像仓库、标签和拉取凭据可分别配置,组件级标签优先于global.imageTag。 - 新增
global.imageSource,默认upstream;设为cn时切换受支持的上游镜像来源。企业版ghcr.io/dynamia-ai/hami-enterprise镜像不会因此改写,非空的global.imageRegistry优先。 - 同 Pod 的 kube-scheduler 通过独立 loopback 端点调用 extender,默认端口
9444。增加节点锁重试和 extender 请求超时配置。 - 增加调度器 PodDisruptionBudget 配置,仅在启用 leader election 且副本数大于
1时创建。 - 增加 monitor 健康探针,默认启用;MIG 实例监控权限由
devicePlugin.monitor.migInstanceAccess控制,默认false。 - 主 Chart 不再包含 HAMi-DRA 子 Chart;
dra、hami-dra配置会被拒绝。本版本使用 HAMi scheduler 与设备插件的部署路径。
升级说明
迁移 r5 的现有部署
厂商字段迁移、旧 Values 整理和渲染校验可参考上游 v2.10 → v2.11.0 升级指南。执行升级时使用企业版 2.11.0-r0 Chart 和 v2.11.0-r0 镜像;镜像标签、以下企业版默认值及配套组件以本发布说明和交付清单为准。
- 导出 r5 的用户 Values,按新路径整理所有仍需保留的覆盖,再使用
--reset-values应用新默认值。不要使用--reuse-values或--reset-then-reuse-values,以免重新带入已删除字段。 - 删除整个
dra、hami-dra配置段,包括原有的dra.enabled: false。企业版不再提供这些安装入口。 - r5 已使用的 HCU 资源名、
devices.ascend.deviceSplitCount等有效配置继续保留,按实际设备插件注册的资源名核对;不重复执行上游从原始 v2.10 出发的资源重命名。使用历史固定 selector 的部署继续保留legacySelectorLabels: true。 - 使用
device-config.content的部署需单独迁移这份完整配置,并保留与既有分配路径一致的nvidia.useDownward等企业版设置;默认 Downward 路径需设为true。该内容不与devices.*合并;复用设备 ConfigMap 的独立 Ascend 插件需同步核对配置。
| 企业版配置 | 本版默认值 | 升级时的处理 |
|---|---|---|
scheduler.useDownward | true | 保留原分配路径;新增校验的条件见下文。 |
devices.nvidia.limitDelivery | env | 继续通过环境变量传递 NVIDIA 限额,不将上游 file 默认值直接带入。 |
devicePlugin.monitor.disableHAMiCoreForWholeGPU | false | 保留企业版整卡运行库路径,不将上游的不同默认值直接带入。 |
devicePlugin.nvidiaDriverRoot | "" | 保留与现有驱动安装方式匹配的设置,不因上游默认采用 auto 就自动改写。 |
核对企业分配接口与运行库
Kubernetes 1.30 及以上且 scheduler.admissionWebhook.enabled: true 时,主 Chart 自动安装 /downward/validate。自定义准入清单需增加该校验端点并配置其 CA;它检查 NVIDIA Downward 运行时限制与容器请求的边界。更早的 Kubernetes 版本不安装此校验;需要同等资源边界校验时,使用非 Downward 分配路径 scheduler.useDownward: false,并按业务更新流程重新创建受影响的 Pod。
使用外置 kube-scheduler 的部署还需调整 extender 调用:/filter、/bind 默认仅监听 127.0.0.1:9444。通过 scheduler.extender.extraArgs 设置可达的 --extender-bind,同步调整调用端地址和网络访问范围;仅修改 Service 不会暴露这两个端点。
组件升级不会替换现有 Pod 的环境变量,也不会让运行中的进程重新加载运行库。需要使用本版运行库修复或新注入行为的工作负载,应按业务维护流程滚动更新或重新创建。安装包与配套组件清单通过约定的交付渠道获取,也可查看软件中心。
Values 说明
以下参数说明取自 Chart 2.11.0-r0 发布包随附的 README,保留原文。默认配置以该版本的 values.yaml 为准。
展开参数表
Values are listed in the same order as in values.yaml.
| Key | Type | Default | Description |
|---|---|---|---|
| global.imageSource | string | "upstream" | Image source: upstream or cn. Set --set global.imageSource=cn to use Alibaba Cloud for kube-scheduler and the default pinned certgen image, and m.daocloud.io for the upstream HAMi, mock-device-plugin, and Lupine server/client repositories. Other registries/repositories, explicit remote-GPU libImage/sessionImage, and supplied device-config content are unchanged. A non-empty global.imageRegistry disables source selection. Custom certgen pins and explicitly pinned kube-scheduler images are used unchanged. |
| global.imageRegistry | string | "" | Global Docker image registry that replaces the registry of every HAMi image, overriding the per-component *.image.registry settings. Leave empty to keep the per-component registries. |
| global.imagePullSecrets | list | [] | Global Docker image pull secrets added to the scheduler, device plugin, webhook patch job, and remote-gpu Pods; the mock device plugin DaemonSet does not use them. Each entry is the name of an existing docker-registry Secret, e.g. myRegistryKeySecretName. |
| global.imageTag | string | "v2.11.0-r0" | Tag (optionally tag@sha256:...) used by the scheduler extender, device plugin and monitor images when their own *.image.tag is empty. An explicit tag replaces the complete value. A non-empty per-component tag takes precedence. The kube-scheduler, webhook patch, mock device plugin and remote-gpu images do not use it. |
| global.gpuHookPath | string | "/usr/local" | Directory prefix on the node where the NVIDIA vGPU hook files are installed (<gpuHookPath>/vgpu). Only used by the NVIDIA device plugin. |
| global.labels | object | {} | Labels added to the scheduler Deployment and PodDisruptionBudget, the NVIDIA and remote-gpu DaemonSets, and the ServiceMonitors. |
| global.annotations | object | {} | Annotations added to the scheduler Deployment and PodDisruptionBudget and to the NVIDIA and remote-gpu DaemonSets. |
| global.managedNodeSelectorEnable | bool | false | When true, add global.managedNodeSelector to the scheduler pod as NODE_SELECTOR_* environment variables. |
| global.managedNodeSelector | object | {"usage":"gpu"} | Node labels exposed to the scheduler pod when global.managedNodeSelectorEnable is true. Each key/value pair becomes a NODE_SELECTOR_<KEY> environment variable on the scheduler pod (key upper-cased, with -, / and . replaced by _), e.g. usage: gpu becomes NODE_SELECTOR_USAGE=gpu. |
| nameOverride | string | "" | Replaces the chart name in resource names and in the app.kubernetes.io/name label. Resource names use the release name alone when it already contains this name, otherwise <release>-<name>. fullnameOverride takes precedence for resource names. |
| fullnameOverride | string | "" | Fully overrides the names of the resources the chart creates, bypassing the <release>-<chart> naming. Truncated to 63 characters. |
| namespaceOverride | string | "" | Namespace the chart's resources are rendered into instead of the release namespace. The chart does not create it, so it must already exist. |
| platform.openshift | bool | false | Enable when deploying HAMi on OpenShift. Applies restricted security contexts to the scheduler extender and webhook jobs, and omits scheduler.patch.runAsUser so OpenShift can assign job UIDs. |
| platform.openshiftVersion | string | "" | |
| openshift.securityContextConstraints.create | bool | true | Create the named SecurityContextConstraints (SCC) and its use ClusterRole (system:openshift:scc:<name>) when platform.openshift=true and name is non-empty. The created SCC allows privileged containers, hostPath volumes, and the host PID namespace. Set to false to reuse an existing SCC and matching ClusterRole. Must be false for the built-in privileged SCC. |
| openshift.securityContextConstraints.name | string | "hami-device-plugin" | Name of the SCC that enabled NVIDIA and mock device-plugin service accounts are allowed to use when platform.openshift=true. Leave empty to skip SCC creation and binding, and manage SCC access separately. |
| selinux.enabled | bool | false | Add a privileged NVIDIA device-plugin init container to recursively relabel host directories devicePlugin.libPath, devicePlugin.monitor.ctrPath, and /tmp/vgpulock. Creates missing directories and sets permissions to 0777 on devicePlugin.monitor.ctrPath and 1777 on /tmp/vgpulock. Requires devicePlugin.enabled=true. Available on any SELinux-enabled Kubernetes distribution, independently of platform.openshift. Uninstalling does not restore host labels or permissions. |
| selinux.type | string | "container_file_t" | SELinux file type passed to chcon -t for the shared vGPU host directories when selinux.enabled=true. Must be a type supported by the node's SELinux policy. |
| selinux.level | string | "s0" | SELinux security level or range passed to chcon -l for the shared vGPU host directories when selinux.enabled=true, for example s0 or s0:c123,c456. |
| schedulerName | string | "hami-scheduler" | Name of the HAMi scheduler. The admission webhook sets it as schedulerName on Pods that request devices, the kube-scheduler container serves it, and the scheduler extender and kube-scheduler use it as their leader election lease name. |
| podSecurityPolicy.enabled | bool | false | Render a policy/v1beta1 PodSecurityPolicy and its use RBAC rule for the webhook certificate creation and patch jobs. Only takes effect when scheduler.admissionWebhook.enabled=true, scheduler.patch.enabled=true, and scheduler.certManager.enabled=false. Keep disabled on Kubernetes 1.25 and later, where the PodSecurityPolicy API is removed. |
| scheduler.nodeName | string | "" | Node to run the scheduler Pod on, bypassing scheduling, for example when HAMi replaces the default scheduler. Empty lets Kubernetes place it; scheduler.nodeSelector can also be set. |
| scheduler.overwriteEnv | string | "false" | Default for whether the admission webhook hides devices from containers that request none: NVIDIA_VISIBLE_DEVICES=none for NVIDIA and an empty ASCEND_VISIBLE_DEVICES for Ascend. The Pod annotations hami.io/overwrite-env and hami.io/overwrite-env-containers override it. |
| scheduler.defaultSchedulerPolicy.nodeSchedulerPolicy | string | "binpack" | Default node scoring policy: binpack fills busy nodes first, spread prefers idle ones. The hami.io/node-scheduler-policy Pod annotation overrides it. |
| scheduler.defaultSchedulerPolicy.gpuSchedulerPolicy | string | "spread" | Default device scoring policy within a node: binpack, spread, or topology-aware. topology-aware also makes the NVIDIA device plugin report GPU topology scores. The hami.io/gpu-scheduler-policy Pod annotation overrides it. |
| scheduler.defaultSchedulerPolicy.nodeAffinityScoreWeight | float | 0 | Bonus added to the node score for nodes matching a Pod's preferred nodeAffinity, scaled by the matched weight ratio (0..1). 0 disables it. HAMi returns a single node from Filter, so other kube-scheduler score plugins are still ignored. Values above 30 let the preference outweigh GPU usage. |
| scheduler.metricsBindAddress | string | ":9395" | Address the scheduler serves Prometheus metrics on, as [host]:port. The chart also takes the port from it for the metrics container port. |
| scheduler.profilingBindAddress | string | "127.0.0.1:6060" | Address of the pprof server, used only when --profiling is added to scheduler.extender.extraArgs. Keep it on loopback and reach it with kubectl port-forward. |
| scheduler.kubeQPS | string | "" | Queries per second the scheduler may send to kube-apiserver. Empty keeps the client default of 5. Raise it, with kubeBurst, on large clusters. |
| scheduler.kubeBurst | string | "" | Request burst the scheduler may send to kube-apiserver. Empty keeps the client default of 10. |
| scheduler.kubeTimeout | string | "" | Timeout in seconds for scheduler requests to kube-apiserver. Empty or 0 means no timeout. |
| scheduler.nodeLockRetryTimeout | string | "" | How long Bind retries locking a node held by another member of the same PodGroup, as a duration such as 28s. Empty keeps the default of 28s, and 0 disables the retry. Keep it below extenderHTTPTimeout. |
| scheduler.extenderHTTPTimeout | int | 30 | Timeout in seconds that kube-scheduler gives each call to the HAMi extender. Keep it above nodeLockRetryTimeout, or kube-scheduler cancels Bind while the extender is still retrying a node lock. |
| scheduler.extenderPort | int | 9444 | Loopback port the extender serves /filter and /bind on, for the kube-scheduler container in the same Pod. It must differ from scheduler.service.httpTargetPort. If kube-scheduler runs outside the Pod, change --extender-bind through scheduler.extender.extraArgs. |
| scheduler.forceOverwriteDefaultScheduler | bool | true | Let the admission webhook replace default-scheduler with schedulerName on Pods that request devices. Kubernetes fills in default-scheduler for Pods that name no scheduler, so when false the webhook leaves every Pod that does not already name schedulerName unchanged. |
| scheduler.livenessProbe | bool | false | Add liveness probes to the scheduler containers. |
| scheduler.leaderElect | bool | true | Enable leader election for the scheduler extender and kube-scheduler. Required for replicas above 1; it also adds a readiness probe, spreads replicas across nodes when possible, and allows the PodDisruptionBudget. Only the leader receives Service traffic. |
| scheduler.replicas | int | 1 | Number of scheduler replicas. Used only when leaderElect is true; otherwise one replica runs. |
| scheduler.podDisruptionBudget.minAvailable | int | 1 | Minimum available scheduler Pods for the PodDisruptionBudget, which is created only when leaderElect is true and replicas is above 1. Setting podDisruptionBudget.maxUnavailable uses that instead. |
| scheduler.kubeScheduler.enabled | bool | true | Run a kube-scheduler container in the scheduler Pod, configured with schedulerName and the HAMi extender. When false, provide a scheduler named schedulerName that calls the extender yourself; the extender listens on loopback, so also change --extender-bind through scheduler.extender.extraArgs. |
| scheduler.kubeScheduler.image.registry | string | "registry.k8s.io" | Registry of the kube-scheduler image. global.imageRegistry overrides it. global.imageSource=cn selects registry.cn-hangzhou.aliyuncs.com/google_containers/kube-scheduler for tag-based upstream references. |
| scheduler.kubeScheduler.image.repository | string | "kube-scheduler" | Repository of the kube-scheduler image. |
| scheduler.kubeScheduler.image.tag | string | "" | Tag of the kube-scheduler image. Empty uses the cluster's Kubernetes version without distribution suffixes, such as v1.31.1 for v1.31.1+k3s1. It is not pinned by digest automatically. global.imageTag does not apply to it. |
| scheduler.kubeScheduler.image.pullPolicy | string | "IfNotPresent" | Pull policy of the kube-scheduler image. |
| scheduler.kubeScheduler.image.pullSecrets | list | [] | Names of existing image pull Secrets for the scheduler Pod, added after global.imagePullSecrets. Used only when enabled is true. |
| scheduler.kubeScheduler.resources | object | {} | Resource requests and limits for the kube-scheduler container. |
| scheduler.kubeScheduler.extraNewArgs | list | ["--config=/config/config.yaml","-v=4"] | Complete kube-scheduler arguments on Kubernetes 1.22 and later. Keep --config=/config/config.yaml, the chart-generated KubeSchedulerConfiguration that sets schedulerName, leader election, and the HAMi extender. |
| scheduler.kubeScheduler.extraArgs | list | ["--policy-config-file=/config/config.json","-v=4"] | Extra kube-scheduler arguments on Kubernetes versions before 1.22, after the chart's --scheduler-name flag and before its leader election flags. Keep --policy-config-file=/config/config.json, the chart-generated Policy that adds the HAMi extender. |
| scheduler.extender.image.registry | string | "ghcr.io" | Registry of the scheduler extender image. global.imageRegistry overrides it. For docker.io/projecthami/hami, set global.imageSource=cn and leave global.imageRegistry empty to use m.daocloud.io/docker.io/projecthami/hami with the same tag and digest. |
| scheduler.extender.image.repository | string | "dynamia-ai/hami-enterprise" | Repository of the scheduler extender image. |
| scheduler.extender.image.tag | string | "" | Tag of the scheduler extender image. Empty uses global.imageTag. |
| scheduler.extender.image.pullPolicy | string | "IfNotPresent" | Pull policy of the scheduler extender image. |
| scheduler.extender.image.pullSecrets | list | [] | Names of existing image pull Secrets for the scheduler Pod, added after global.imagePullSecrets. |
| scheduler.extender.resources | object | {} | Resource requests and limits for the scheduler extender container. |
| scheduler.extender.extraArgs | list | ["--debug","-v=4"] | Extra command-line flags for the scheduler extender, appended after the flags the chart generates, such as --profiling to serve pprof on scheduler.profilingBindAddress. |
| scheduler.nodeLockExpire | string | "5m" | Age after which a node lock is treated as expired, as a positive duration such as 5m. An unparsable or non-positive value is ignored and 5m is used. A --node-lock-timeout flag in scheduler.extender.extraArgs overrides it. |
| scheduler.podAnnotations | object | {} | Annotations for the scheduler Pods. |
| scheduler.tolerations | list | [] | Tolerations for the scheduler Pods. |
| scheduler.admissionWebhook.enabled | bool | true | Install the mutating admission webhook. It rejects Pods that preset HAMi allocation annotations, and for Pods that request devices it applies each device backend's admission changes, sets schedulerName, checks ResourceQuota, and rejects privileged containers. When false, Pods that should use HAMi must set schedulerName and their device settings themselves, and the scheduler serves its HTTP port without TLS. |
| scheduler.admissionWebhook.customURL.enabled | bool | false | Have kube-apiserver call the webhook at https://<host>:<port><path>, such as a node port, instead of through the scheduler Service. |
| scheduler.admissionWebhook.customURL.host | string | "127.0.0.1" | Host name or IP address in the custom webhook URL. The certificate from the patch Jobs includes it; the cert-manager certificate does not. |
| scheduler.admissionWebhook.customURL.port | int | 31998 | Port in the custom webhook URL. The default matches scheduler.service.schedulerPort. |
| scheduler.admissionWebhook.customURL.path | string | "/webhook" | Path in the custom webhook URL. The scheduler serves the webhook on /webhook. |
| scheduler.admissionWebhook.whitelistNamespaces | list | unset | Names of namespaces whose Pods the webhook skips, such as kube-system. Applies only when manageNamespaceSelector is true. |
| scheduler.admissionWebhook.manageNamespaceSelector | bool | true | Render the webhook's namespaceSelector. Set it to false on platforms that manage that field themselves, such as AKS, to avoid server-side apply conflicts. whitelistNamespaces and namespaceSelector then have no effect. |
| scheduler.admissionWebhook.namespaceSelector.matchLabels | object | unset | Labels a namespace must have for the webhook to apply to its Pods. Applies only when manageNamespaceSelector is true. |
| scheduler.admissionWebhook.namespaceSelector.matchExpressions | list | [] | Extra namespace selector expressions, added after the built-in one that skips namespaces labeled hami.io/webhook: ignore. Applies only when manageNamespaceSelector is true. |
| scheduler.admissionWebhook.objectSelector.matchExpressions | list | [] | Extra Pod selector expressions, added after the built-in one that skips Pods labeled hami.io/webhook: ignore. Labels a Pod must have can also be set as objectSelector.matchLabels. |
| scheduler.admissionWebhook.reinvocationPolicy | string | "Never" | Whether kube-apiserver calls the webhook again after a later mutating webhook changes the Pod: Never or IfNeeded. |
| scheduler.admissionWebhook.failurePolicy | string | "Ignore" | What kube-apiserver does when it cannot reach the webhook. Ignore admits the Pod without HAMi's changes and without an error. Fail rejects the request; because the webhook handles every Pod create and update in the selected namespaces, an unavailable scheduler then blocks them all. |
| scheduler.certManager.enabled | bool | false | Use cert-manager for the admission webhook certificate. The chart creates a self-signed Issuer and a Certificate for <scheduler Service>.<namespace>.svc stored in Secret <fullname>-scheduler-tls, and annotates the webhook so cert-manager injects its CA. Requires cert-manager in the cluster and admissionWebhook.enabled. Takes precedence over patch. With both certManager.enabled and patch.enabled false, create that Secret with tls.crt and tls.key and set the webhook CA yourself. |
| scheduler.patch.enabled | bool | true | Use kube-webhook-certgen Jobs for the admission webhook certificate when certManager.enabled is false. A pre-install and pre-upgrade hook Job writes a self-signed certificate to Secret <fullname>-scheduler-tls, and a post-install and post-upgrade hook Job sets its CA on the webhook. Requires admissionWebhook.enabled. |
| scheduler.patch.imageNew.registry | string | "registry.k8s.io" | Registry of the certgen image. If registry.k8s.io is inaccessible, set global.imageSource=cn before installing or upgrading. For a manual mirror override, use registry.cn-hangzhou.aliyuncs.com with the matching repository and tag below. global.imageRegistry overrides this field; leave it empty to use this per-component setting. |
| scheduler.patch.imageNew.repository | string | "ingress-nginx/kube-webhook-certgen" | Repository of the certgen image. For the China mirror, use google_containers/kube-webhook-certgen. |
| scheduler.patch.imageNew.tag | string | "v1.6.9@sha256:01038e7de14b78d702d2849c3aad72fd25903c4765af63cf16aa3398f5d5f2dd" | Certgen version and multi-architecture index digest, in tag@sha256:... format. For the China mirror, use v1.6.9@sha256:78bf4f774a27f29a5e6333c11708ac4645139e7bb2b74aa9e9c4574935f7a989. global.imageSource=cn switches the default pinned reference to the mirror automatically. Source and mirror index digests differ; for manual overrides, change registry, repository, and tag together. |
| scheduler.patch.imageNew.pullPolicy | string | "IfNotPresent" | Pull policy of the certgen image. |
| scheduler.patch.imageNew.pullSecrets | list | [] | Names of existing image pull Secrets for the certgen Jobs, added after global.imagePullSecrets. |
| scheduler.patch.priorityClassName | string | "" | PriorityClass for the certgen Job Pods. Empty sets none. |
| scheduler.patch.podAnnotations | object | {} | Annotations for the certgen Job Pods. |
| scheduler.patch.nodeSelector | object | {} | Node selector for the certgen Job Pods. |
| scheduler.patch.tolerations | list | [] | Tolerations for the certgen Job Pods. |
| scheduler.patch.runAsUser | int | 2000 | User ID the certgen Job Pods run as. They always run as non-root. Not set when platform.openshift is true, so the OpenShift SCC can assign an allowed UID, or when this value is 0 or empty. |
| scheduler.service.type | string | "NodePort" | Type of the scheduler Service, which routes only to the scheduler Pod holding leadership. With NodePort, the chart also sets the node ports from schedulerPort and monitorPort. Empty uses ClusterIP. |
| scheduler.service.httpPort | int | 443 | Service port for the scheduler's HTTP port. The admission webhook and the device plugin's default NUMA refit endpoint reach the scheduler through it. |
| scheduler.service.schedulerPort | int | 31998 | Node port for the HTTP port when type is NodePort. |
| scheduler.service.monitorPort | int | 31993 | Service port for scheduler metrics. When type is NodePort, it is also used as the node port, so it must be in the cluster's node port range. |
| scheduler.service.monitorTargetPort | int | 9395 | Container port the metrics Service port targets, by number or name. The default metrics names the container port taken from scheduler.metricsBindAddress. |
| scheduler.service.httpTargetPort | int | 9443 | Container port the scheduler serves the admission webhook, /refit, and probes on. It must differ from scheduler.extenderPort. |
| scheduler.service.labels | object | {} | Extra labels for the scheduler Service. |
| scheduler.service.annotations | object | {} | Annotations for the scheduler Service. |
| scheduler.networkPolicy.enabled | bool | false | Create a NetworkPolicy for the scheduler Pods. It admits traffic to the HTTP port (scheduler.service.httpTargetPort), which serves the admission webhook and the device plugin's /refit calls, only from this release's device plugin Pods and the kube-system namespace, leaves the metrics port open to all sources, and denies other ingress. Enable it only if your CNI matches kube-apiserver traffic against namespaceSelector: when kube-apiserver uses host networking, many CNIs do not, and with admissionWebhook.failurePolicy: Ignore Pods are then admitted without HAMi's changes and without an error. |
| scheduler.useDownward | bool | true | |
| devicePlugin.enabled | bool | true | Deploy the NVIDIA device plugin DaemonSet, including the vGPU monitor container. |
| devicePlugin.gpuOperatorToolkitReady.enabled | bool | false | Wait for <hostPath>/toolkit-ready in an init container before starting the device plugin and monitor. |
| devicePlugin.gpuOperatorToolkitReady.hostPath | string | "/run/nvidia/validations" | Host directory containing toolkit-ready. Also supplies driver-ready when devicePlugin.nvidiaDriverRoot=auto. |
| devicePlugin.gpuOperatorToolkitReady.securityContext | object | {"privileged":true,"runAsUser":0} | Security context for the toolkit-validation init container. |
| devicePlugin.image.registry | string | "ghcr.io" | NVIDIA device plugin image registry. Overridden by a non-empty global.imageRegistry. For docker.io/projecthami/hami, set global.imageSource=cn and leave global.imageRegistry empty to use m.daocloud.io/docker.io/projecthami/hami with the same tag and digest. |
| devicePlugin.image.repository | string | "dynamia-ai/hami-enterprise" | Image repository for the NVIDIA device plugin container and the selinux-relabel and toolkit-validation init containers when enabled. |
| devicePlugin.image.tag | string | "" | NVIDIA device plugin image tag. Empty uses global.imageTag. |
| devicePlugin.image.pullPolicy | string | "IfNotPresent" | Image pull policy for the NVIDIA device plugin and its enabled init containers: Always, IfNotPresent, or Never. |
| devicePlugin.image.pullSecrets | list | [] | Image pull Secret names for the device plugin Pod, appended to global.imagePullSecrets, e.g. [myRegistryKeySecretName]. |
| devicePlugin.vgpuCache.gracePeriod | string | "5m" | Minimum cache directory age, measured from its modification time, before garbage collection. Use Go duration syntax such as 30s, 5m, or 1h. Empty or invalid values use 5m; negative durations prevent startup. 0s removes the age grace period but still requires live Pod confirmation. Independent of monitor resync and the GC scan interval. |
| devicePlugin.monitor.image.registry | string | "ghcr.io" | Monitor image registry. Overridden by a non-empty global.imageRegistry. For docker.io/projecthami/hami, set global.imageSource=cn and leave global.imageRegistry empty to use m.daocloud.io/docker.io/projecthami/hami with the same tag and digest. |
| devicePlugin.monitor.image.repository | string | "dynamia-ai/hami-enterprise" | Monitor image repository. |
| devicePlugin.monitor.image.tag | string | "" | Monitor image tag. Empty uses global.imageTag. |
| devicePlugin.monitor.image.pullPolicy | string | "IfNotPresent" | Monitor image pull policy: Always, IfNotPresent, or Never. |
| devicePlugin.monitor.image.pullSecrets | list | [] | Monitor image pull Secret names. Currently unused by the DaemonSet; configure devicePlugin.image.pullSecrets or global.imagePullSecrets for the shared Pod. |
| devicePlugin.monitor.ctrPath | string | "/usr/local/vgpu/containers" | Shared per-container libvgpu cache path used by the device plugin and monitor. |
| devicePlugin.monitor.metricsBindAddress | string | ":9394" | Address the monitor serves /metrics on. The container port, Service target port and probes follow it. Loopback addresses are rejected. Set it here, not in extraArgs. |
| devicePlugin.monitor.probes.enabled | bool | true | Run the monitor liveness and readiness probes. Disable to debug inside the container. |
| devicePlugin.monitor.resyncInterval | string | "5m" | Monitor Pod informer resync interval and grace period for releasing mappings of missing Pods. Does not control directory GC or Prometheus scrape frequency. |
| devicePlugin.monitor.disableHAMiCoreForWholeGPU | bool | false | Disable the HAMi-core hook for whole-GPU containers when memory and core scaling are both 1; the monitor reads their metrics from NVML instead. Whole-GPU means a non-MIG device with full device memory allocated and cores unrequested or gpucores=100. |
| devicePlugin.monitor.extraArgs | list | ["-v=4"] | Additional monitor command-line arguments as a list of strings. |
| devicePlugin.monitor.extraEnvs | list | [] | Extra Kubernetes env entries for the vGPU monitor, as a list of objects with name and value or valueFrom. @schema type: [array, object] |
| devicePlugin.monitor.securityContext | object | {"allowPrivilegeEscalation":false,"capabilities":{"add":["SYS_ADMIN"],"drop":["ALL"]}} | Security context for the monitor container. |
| devicePlugin.monitor.migInstanceAccess | bool | false | MIG instance monitoring requires privileged access for the monitor container; if you don't use MIG, you can set this to false. |
| devicePlugin.monitor.resources | object | {} | Resource requests and limits for the monitor container. |
| devicePlugin.numaRefit.enabled | bool | false | NUMA alignment refit (#2080): when enabled, the device plugin asks the scheduler to re-run its fit over the NUMA-restricted device set kubelet allows, keeping scheduler accounting authoritative. Requires devices.nvidia.enableNumaTopology plus the per-node enablegetpreferredallocation node configuration; pods opt in with the hami.io/numa-alignment annotation (best-effort or strict). Disabled by default: mismatches are only logged. |
| devicePlugin.numaRefit.schedulerEndpoint | string | "" | Scheduler base URL override. Defaults to the in-cluster scheduler service, https://<release>-scheduler.<namespace>.svc:<scheduler.service.httpPort>. |
| devicePlugin.numaRefit.tlsInsecure | bool | false | Must remain false: refit sends a ServiceAccount token and requires TLS certificate verification. Enabling refit requires caFile or caSecret containing the CA that signed the scheduler serving certificate. |
| devicePlugin.numaRefit.caFile | string | "" | Path to a CA bundle already present inside the device-plugin container. |
| devicePlugin.numaRefit.caSecret | string | "" | Name of a Secret (key ca.crt) to mount as the CA bundle instead; used when caFile is empty. |
| devicePlugin.reportNodeCapacity | bool | false | Report GPU memory and core resources in Node capacity and allocatable. The reportnodecapacity JSON field in nodeConfiguration.config overrides this setting for matching nodes. |
| devicePlugin.nodeConfiguration.config | string | Example node configuration; see values.yaml. | JSON content for the generated device plugin ConfigMap key config.json; ignored when externalConfigName is set. The nodeconfig list selects entries whose name matches the node. If none match, it selects matching nodelabelselector entries; if neither matches, it uses entries named *. For example, use {"nodelabelselector":{"matchLabels":{"gpu.example.com/pool":"mig"}},"operatingmode":"mig"}. Node labels are read at plugin startup; restart the plugin after changing them and use labels available when the node registers. Selected entries apply in list order; later overrides take precedence. For per-node resource settings, null, zero, or false does not replace an existing non-zero or true value. Empty operatingmode and filterdevices values are ignored. Entries with both or neither of name and nodelabelselector, or with an empty or invalid label selector, are skipped. |
| devicePlugin.nodeConfiguration.externalConfigName | string | "" | Existing ConfigMap containing a config.json key in the Pod namespace. When non-empty, the chart mounts it instead of creating a ConfigMap from nodeConfiguration.config. |
| devicePlugin.migStrategy | string | "none" | Strategy for exposing NVIDIA MIG devices: none, single, or mixed. mixed exposes MIG profiles as separate resources. Passed as --mig-strategy. |
| devicePlugin.disablecorelimit | string | "false" | Set to "true" to disable HAMi-core utilization limits. This value is merged with nvidia.disableCoreLimit in device-config.yaml; either one being true disables limits, while "false" does not override a true file value. |
| devicePlugin.passDeviceSpecsEnabled | bool | true | Pass device specifications to kubelet in Allocate responses. Ignored when deviceListStrategy contains only CDI strategies. Null leaves the plugin default unchanged (disabled). |
| devicePlugin.deviceListStrategy | string | "envvar" | Comma-separated device list strategies passed to the container runtime: envvar, volume-mounts, cdi-annotations, or cdi-cri. CDI strategies require a CDI-capable runtime; cdi-cri also requires kubelet/CRI support for CDI devices. |
| devicePlugin.nvidiaHookPath | string | "" | Path for NVIDIA CDI hooks in generated CDI specifications. Null omits NVIDIA_CDI_HOOK_PATH and keeps the plugin default /usr/bin/nvidia-ctk. |
| devicePlugin.nvidiaDriverRoot | string | "" | Root of the NVIDIA driver on the host. Set to "auto" to read the driver and device roots from GPU Operator's driver-ready contract. If the contract is absent, both roots default to / for a host-installed driver. In auto mode the device plugin uses the driver injected by the NVIDIA Container Toolkit, so it must run with the NVIDIA runtime; the host root is mounted read-only only when deviceListStrategy uses a cdi-* strategy. |
| devicePlugin.gdrcopyEnabled | bool | false | Enable GDRCopy support for allocated GPU containers. A boolean sets GDRCOPY_ENABLED; null leaves the plugin default unchanged (disabled). Requires the corresponding host drivers and runtime support. |
| devicePlugin.gdsEnabled | bool | false | Enable GPUDirect Storage support for allocated GPU containers. A boolean sets GDS_ENABLED; null leaves the plugin default unchanged (disabled). Requires the corresponding host drivers and runtime support. |
| devicePlugin.mofedEnabled | bool | false | Enable MOFED support for allocated GPU containers. A boolean sets MOFED_ENABLED; null leaves the plugin default unchanged (disabled). Requires the corresponding host drivers and runtime support. |
| devicePlugin.extraArgs | list | ["-v=4"] | Additional NVIDIA device plugin command-line arguments as a list of strings. |
| devicePlugin.extraEnvs | list | [] | Extra Kubernetes env entries for the device plugin, as a list of objects with name and value or valueFrom. @schema type: [array, object] |
| devicePlugin.service.type | string | "NodePort" | Service type for vGPU monitor metrics. Empty uses NodePort; ClusterIP exposes metrics only within the cluster. |
| devicePlugin.service.httpPort | int | 31992 | Monitor Service port. Also used as the nodePort when service.type=NodePort. Empty or zero uses 31992; the target port follows monitor.metricsBindAddress. |
| devicePlugin.service.labels | object | {} | Additional labels for the vGPU monitor Service, as a map of label names to string values. |
| devicePlugin.service.annotations | object | {} | Annotations for the vGPU monitor Service, as a map of annotation names to string values. |
| devicePlugin.pluginPath | string | "/var/lib/kubelet/device-plugins" | Host kubelet device-plugin socket directory, mounted at /var/lib/kubelet/device-plugins inside the device plugin container. |
| devicePlugin.libPath | string | "/usr/local/vgpu" | Host directory for HAMi-core library files, mounted at <global.gpuHookPath>/vgpu inside the device plugin container. |
| devicePlugin.podAnnotations | object | {} | Additional annotations for NVIDIA device plugin Pods, as a map of annotation names to string values. |
| devicePlugin.hostPID | bool | true | Use the host PID namespace for NVIDIA device plugin Pods. Must be true when hostPIDBroker.enabled is true. |
| devicePlugin.hostPIDBroker.enabled | bool | false | Enable the host PID broker so HAMi-core can query the device plugin for its host PID. Requires hostPID=true; mounts the broker socket directory at /var/run/hami/hostpid. See Host PID broker. |
| devicePlugin.hostNetwork | bool | false | Use the host network for NVIDIA device plugin Pods. When enabled, the chart sets dnsPolicy: ClusterFirstWithHostNet so the scheduler Service remains resolvable. |
| devicePlugin.securityContext | object | {"allowPrivilegeEscalation":true,"capabilities":{"add":["SYS_ADMIN"],"drop":["ALL"]},"privileged":true} | Kubernetes container security context for the NVIDIA device plugin. |
| devicePlugin.nvidiaNodeSelector | object | {"gpu":"on"} | Node labels required for NVIDIA device plugin Pods, as a map of label names to string values. |
| devicePlugin.tolerations | list | [{"effect":"NoSchedule","key":"nvidia.com/gpu","operator":"Exists"}] | Kubernetes tolerations for NVIDIA device plugin Pods, as a list of objects with key, operator, value, effect, and optional tolerationSeconds. |
| devicePlugin.updateStrategy | object | {"rollingUpdate":{"maxUnavailable":1},"type":"RollingUpdate"} | Kubernetes DaemonSet update strategy: RollingUpdate or OnDelete. For RollingUpdate, configure rollingUpdate.maxUnavailable as a count or percentage. |
| devicePlugin.resources | object | {} | Kubernetes resource requests and limits for the NVIDIA device plugin container, for example {requests: {cpu: 100m, memory: 100Mi}, limits: {memory: 1Gi}}. |
| devicePlugin.cdiVendor | string | "" | |
| mockDevicePlugin.enabled | bool | false | Deploy the mock device plugin as a DaemonSet on all eligible nodes without node selectors. |
| mockDevicePlugin.image.registry | string | "docker.io" | Image registry, overridden by global.imageRegistry. |
| mockDevicePlugin.image.repository | string | "projecthami/mock-device-plugin" | Image repository for the mock-device-plugin. |
| mockDevicePlugin.image.tag | string | "1.0.1@sha256:61d08a5641251ed776e4701c0dcf05c2c2bed5a58b39b8a11ec85ca5e57e674a" | Image tag (required), pinned by digest; does not inherit global.imageTag. |
| mockDevicePlugin.image.pullPolicy | string | "IfNotPresent" | Image pull policy: Always, IfNotPresent, or Never. |
| devices.metax.resourceCountName | string | "metax-tech.com/gpu" | Resource name for non-negative integer whole-GPU counts. Must match the vendor device plugin; 0 requests no device. |
| devices.metax.resourceVCountName | string | "metax-tech.com/sgpu" | Resource name for non-negative integer sGPU counts in container limits. Required for sGPU memory or core requests; 0 requests no device. See MetaX workload examples. |
| devices.metax.resourceVCoreName | string | "metax-tech.com/vcore" | Resource name for each sGPU's requested compute percentage (integer, 0-100; 40 means 40%). Omission requests 100; 0 reserves no compute. Pods annotated with metax-tech.com/sgpu-app-class: online do not reserve compute through this resource. |
| devices.metax.resourceVMemoryName | string | "metax-tech.com/vmemory" | Resource name for the memory limit on each sGPU. Bare integers are in GiB; use a quantity such as 2048Mi to specify the unit explicitly. An omitted or zero limit requests the GPU's full memory. |
| devices.metax.sgpuTopologyAware | bool | false | Enable topology-aware node scoring for sGPU requests of 100% compute. Does not apply to online-class Pods or whole-GPU requests. The Pod annotation metax-tech.com/sgpu-topology-aware overrides this setting when it contains a valid boolean. @schema type: [boolean, string] |
| devices.hygon.memoryFactor | int | 1 | Non-negative integer multiplier for explicit HCU memory requests in MiB and their ResourceQuota limits. For example, 1024 makes a memory request of 4 mean 4 GiB. 0 and 1 leave requests unscaled. |
| devices.hygon.resourceCountName | string | "hygon.com/hcunum" | Resource name for non-negative integer vHCU counts. Required for HCU memory or core requests; 0 requests no device. Must match the vendor device plugin. See HCU workload examples. |
| devices.hygon.resourceMemoryName | string | "hygon.com/hcumem" | Resource name for the memory requested on each vHCU, in integer MiB units, scaled by devices.hygon.memoryFactor when it is greater than 1. Omission or 0 requests the selected HCU's full memory. |
| devices.hygon.resourceCoreName | string | "hygon.com/hcucores" | Resource name for each vHCU's requested compute percentage (integer, 0-100; 40 means 40%). Omission or 0 reserves no compute. An explicit 100 requests the card exclusively. |
| devices.cambricon.resourceCountName | string | "cambricon.com/vmlu" | Resource name for the number of MLUs a container requests; it must match the Cambricon device plugin. A zero count requests no device; HAMi reads memory and core requests only for a positive count. Sharing needs one MLU and the plugin in dynamic-smlu mode; a count above 1 allocates whole MLUs and ignores the memory and core values. A non-empty device-config.content or a bundled files/device-config.yaml replaces the generated device configuration, so set the name there too. |
| devices.cambricon.resourceMemoryName | string | "cambricon.com/mlu.smlu.vmemory" | Resource name for MLU memory, requested in integer units of 256 MiB (20 is 5120 MiB); run the plugin with min-dsmlu-unit=256. Omitted or 0 reserves the whole MLU memory. Set the name in device-config.content or files/device-config.yaml too when either replaces the generated configuration. |
| devices.cambricon.resourceCoreName | string | "cambricon.com/mlu.smlu.vcore" | Resource name for MLU compute, requested as an integer percentage from 0 to 100; omitted requests 100. Set the name in device-config.content or files/device-config.yaml too when either replaces the generated configuration. |
| devices.amd.resourceCountName | string | "amd.com/gpu" | Resource name for AMD GPU count requests. Must match the resource advertised by the AMD device plugin; 0 requests no device. An empty value skips the node GPU health check. |
| devices.amd.resourceMemoryName | string | "amd.com/gpumem" | Resource name for per-GPU memory requests in MiB. defaultMemory applies when neither memory resource is specified. |
| devices.amd.resourceMemoryPercentageName | string | "amd.com/gpumem-percentage" | Resource name for requesting GPU memory as a percentage (0-100). If both a positive fixed memory size and a percentage are set, the fixed size takes priority. If either value is explicitly set to 0 and the other is not positive, the full GPU memory is used. If neither is set, defaultMemory is used. An empty value disables percentage memory requests. |
| devices.amd.resourceCoreName | string | "amd.com/gpucores" | Resource name for per-GPU core percentage requests (1-100). Omitted uses defaultCores, or all cores when that value is 0. |
| devices.amd.runtimeClassName | string | "" | RuntimeClass the admission webhook adds to AMD GPU Pods that do not already specify one. The chart does not create it; empty leaves the Pod unchanged. |
| devices.amd.defaultGPUNum | int | 1 | Default GPU count added by the admission webhook when memory, memory percentage, or cores are requested without a GPU count. Values below 1 are treated as 1. A GPU count already specified in resource requests is preserved. |
| devices.amd.defaultMemory | int | 0 | Default memory per requested GPU in MiB. Used when neither memory resource is specified; 0 uses the full GPU memory. |
| devices.amd.defaultCores | int | 0 | Default core percentage per requested GPU. Used when no core request is specified; 0 uses all cores. Values are clamped to 0-100. |
| devices.amd.customresources | list | ["amd.com/gpu","amd.com/gpumem","amd.com/gpucores"] | Resource names managed by the scheduler extender. Include resourceCountName, resourceMemoryName, and resourceCoreName; they are not added automatically. resourceMemoryPercentageName is added automatically. |
| devices.awsneuron.resourceCountName | string | "aws.amazon.com/neuron" | Resource name for positive integer counts of whole Neuron devices. Takes precedence over devices.awsneuron.resourceCoreName requests when both are present. |
| devices.awsneuron.resourceCoreName | string | "aws.amazon.com/neuroncore" | Resource name for positive integer NeuronCore counts. Cores per device depend on the node; one core is not always half a device. See the NeuronCore workload example. |
| devices.awsneuron.customresources | list | ["aws.amazon.com/neuron","aws.amazon.com/neuroncore"] | List of resource-name strings managed by the scheduler extender. Include the names configured by devices.awsneuron.resourceCountName and devices.awsneuron.resourceCoreName; they are not added automatically. |
| devices.kunlun.resourceCountName | string | "kunlunxin.com/xpu" | Resource name for non-negative integer physical XPU counts. Must match the device plugin and Pod resources. See Kunlunxin scheduling. |
| devices.kunlun.resourceVCountName | string | "kunlunxin.com/vxpu" | Resource name for non-negative integer virtual XPU counts. Use with devices.kunlun.resourceVMemoryName for memory partitions. See VXPU usage. |
| devices.kunlun.resourceVMemoryName | string | "kunlunxin.com/vxpu-memory" | Resource name for virtual XPU memory requests in integer MiB per XPU. Requires a virtual XPU count request. See memory partition sizes. |
| devices.kunlun.enabled | bool | true | Add Kunlun resource names to the bundled kube-scheduler's extender configuration. Setting false retains the device configuration. The vendor device plugin must be installed separately. |
| devices.kunlun.customresources | list | [] | Additional resource-name strings managed by the scheduler extender when devices.kunlun.enabled is true. Standard names are added automatically from the resource fields. |
| devices.enflame.resourceNameGCU | string | "enflame.com/gcu" | Resource name for non-negative integer whole-GCU counts. Must match the device plugin and Pod resources. |
| devices.enflame.resourceNameDRSGCU | string | "enflame.com/drs-gcu" | Resource name for a positive integer DRS slice count on one GCU. Takes precedence over memory/core requests. See DRS workload examples. |
| devices.enflame.resourceNameGCUMemory | string | "enflame.com/gcu-memory" | Resource name for positive integer memory requests used to select a GCU partition, optionally combined with devices.enflame.resourceNameGCUCore. Documented requests use MiB (20480 = 20 GiB); small values may be interpreted as GiB depending on the available partitions. See Enflame DRS workload examples. |
| devices.enflame.resourceNameGCUCore | string | "enflame.com/gcu-core" | Resource name for a minimum core percentage (integer, 0-100; 40 means 40%), optionally combined with devices.enflame.resourceNameGCUMemory. Zero or omission sets no core minimum. See DRS workload examples. |
| devices.enflame.enabled | bool | true | Add Enflame resource names to the bundled kube-scheduler's extender configuration. Setting false retains the device configuration. The vendor device plugin must be installed separately. |
| devices.enflame.customresources | list | [] | Additional resource-name strings managed by the scheduler extender when devices.enflame.enabled is true. Standard names are added automatically from the resource fields. |
| devices.enflame.resourceNameVGCU | string | "enflame.com/vgcu" | |
| devices.enflame.resourceNameVGCUPercentage | string | "enflame.com/vgcu-percentage" | |
| devices.mthreads.enabled | bool | true | Register customresources with the HAMi scheduler extender. The MThreads toolkit must be installed separately. |
| devices.mthreads.customresources | list | ["mthreads.com/vgpu"] | Resource names managed by the scheduler extender when enabled. Include resourceCountName; it is not added automatically. |
| devices.mthreads.resourceCountName | string | "mthreads.com/vgpu" | Resource name for MThreads GPU count requests. Must match the vendor device plugin and the entry in customresources. Requesting only this resource allocates exclusive GPUs. |
| devices.mthreads.resourceMemoryName | string | "mthreads.com/sgpu-memory" | Resource name for sGPU memory requests in integer units of 512 MiB (32 = 16 GiB). When memory is not requested, HAMi uses the selected node's per-card capacity. |
| devices.mthreads.resourceCoreName | string | "mthreads.com/sgpu-core" | Resource name for integer sGPU time-sharing weight requests. HAMi accounts for 16 units per card: a single-card request of 8 consumes 8/16 of its allocatable compute units. Device sharing for multi-cards is not supported. For multi-card requests, HAMi overwrites the compute request with 16 units per card. Actual GPU time depends on the vendor scheduling policy and competing workloads. |
| devices.mthreads.memoryPerCard | list | [96] | Per-card capacities in 512 MiB units used to validate explicit single-device memory limits, not to set node capacity. 96 = MTT S4000 (48 GiB); 160 = MTT S5000 (80 GiB). Use [96, 160] for mixed fleets. Allowed limits are the union of powers of two starting at 2 and each full-card capacity. An empty list falls back to [96]; a legacy scalar is converted to a one-item list. |
| devices.vastai.resourceCountName | string | "vastaitech.com/va" | Resource name for non-negative integer Vastai device counts (cards or dies, depending on the device plugin mode). Must match the device plugin and Pod resources. See Vastai usage. |
| devices.vastai.enabled | bool | true | Add Vastai resource names to the bundled kube-scheduler's extender configuration. Setting false retains the device configuration. The vendor device plugin must be installed separately. |
| devices.vastai.customresources | list | [] | Additional resource-name strings managed by the scheduler extender when devices.vastai.enabled is true. Standard names are added automatically from the resource fields. |
| devices.biren.resourceCountName | string | "birentech.com/gpu" | Resource name for non-negative integer Biren device counts. Must match the device plugin and Pod resources. See Biren usage. |
| devices.biren.enabled | bool | true | Add Biren resource names to the bundled kube-scheduler's extender configuration. Setting false retains the device configuration. The vendor device plugin must be installed separately. |
| devices.biren.customresources | list | [] | Additional resource-name strings managed by the scheduler extender when devices.biren.enabled is true. Standard names are added automatically from the resource fields. |
| devices.nvidia.adoptExistingMIGInstances | bool | false | In the mig operating mode, let the device plugin adopt, when it starts, MIG instances HAMi has no record of creating, so it can reuse or destroy them. When false, they are left untouched. Enable it only when HAMi alone manages the MIG layout of the node. |
| devices.nvidia.migProfileAllowlist | list | Profiles for A30, A100, H100, H20, GH200, H200, B200, and RTX PRO 6000 Blackwell Server Edition | MIG profiles HAMi may create on nodes whose device plugin runs in the mig operating mode. Each entry has models, a list of GPU model names matched as substrings of the name NVML reports, and profiles, a list of MIG profile names such as 1g.10gb, for example {models: ["A30"], profiles: ["1g.6gb", "2g.12gb"]}. The first entry that matches a GPU's model decides the profiles for that GPU; a GPU that no entry matches is not registered. The nvidia.com/mig-profile-preference Pod annotation accepts only listed profiles or their slice prefix, such as 4g. The scheduler and device plugin fail to start if an entry has empty models or profiles, or if a profile name has no . or its part before the first . is not at least two characters ending in g. The part after the . is not checked. A profile that passes this check but names a slice count the GPU does not support, such as 5g.20gb, is skipped on the node. Setting this value replaces the whole default list. |
| devices.nvidia.defaultMemory | int | 0 | GPU memory in MiB assigned to each requested GPU when a container sets neither the resourceMemoryName nor the resourceMemoryPercentageName resource, or sets them to 0. 0 assigns the whole memory of the GPU. memoryFactor does not apply to this value. Must not be negative. |
| devices.nvidia.defaultCores | int | 0 | Percentage of GPU cores, from 0 to 100, assigned to each requested GPU when a container does not set the resourceCoreName resource. 0 reserves no cores and sets no core limit. When the admission webhook is enabled, a container that requests GPUs without a memory amount, or with 100% of the memory, gets 100 instead, which requests each GPU exclusively when deviceCoreScaling is 1. |
| devices.nvidia.defaultGPUNum | int | 1 | Number of GPUs the admission webhook adds to a container that sets a resourceMemoryName, resourceMemoryPercentageName, or resourceCoreName limit but no resourceCountName limit. 0 adds nothing, so such a container is not treated as an NVIDIA GPU request. |
| devices.nvidia.memoryFactor | int | 1 | Multiplier applied to the resourceMemoryName value a container requests, which is otherwise in MiB. For example, 1024 makes nvidia.com/gpumem: 4 request 4 GiB. ResourceQuota limits for that resource are scaled the same way. 0 and 1 disable scaling. Must not be negative. |
| devices.nvidia.resourceCountName | string | "nvidia.com/gpu" | Extended resource name for the number of NVIDIA GPUs a container requests. The device plugin registers it with kubelet, advertising each physical GPU deviceSplitCount times. |
| devices.nvidia.resourceMemoryName | string | "nvidia.com/gpumem" | Extended resource name for the GPU memory a container requests on each GPU, in MiB multiplied by memoryFactor. |
| devices.nvidia.resourceMemoryPercentageName | string | "nvidia.com/gpumem-percentage" | Extended resource name for the GPU memory a container requests on each GPU as a percentage of that GPU's memory, an integer from 0 to 100. 0 is treated as unset. |
| devices.nvidia.resourceCoreName | string | "nvidia.com/gpucores" | Extended resource name for the GPU cores a container requests on each GPU, as an integer percentage from 0 to 100. 0 reserves no cores and sets no core limit. With deviceCoreScaling: 1, 100 requests each GPU exclusively. |
| devices.nvidia.resourcePriorityName | string | "nvidia.com/priority" | Extended resource name for the task priority of a container. The admission webhook passes its limit to HAMi-core as CUDA_TASK_PRIORITY. A lower value is a higher priority, and HAMi-core uses 1 when it is unset. On a shared GPU, vGPUmonitor pauses kernel launches of lower-priority tasks while higher-priority tasks are active, but only for tasks with a resourceCoreName limit between 1 and 99 whose core limit is being enforced (see gpuCorePolicy). |
| devices.nvidia.deviceSplitCount | int | 10 | Number of vGPUs each physical GPU is advertised as, which is the maximum number of containers that can share one GPU. In the mig operating mode, the number of MIG instances a GPU can hold limits sharing instead. A device-plugin nodeconfig entry can override it per node with devicesplitcount. |
| devices.nvidia.deviceMemoryScaling | float | 1 | Factor applied to the memory each GPU reports to the scheduler. Accepts decimals, such as 1.5. A value above 1 lets the memory requests on one GPU add up to more than its physical memory, and the device plugin sets CUDA_OVERSUBSCRIBE=true in those containers. Not applied in the mig operating mode. A device-plugin nodeconfig entry can override it per node with devicememoryscaling. @schema type: number |
| devices.nvidia.deviceCoreScaling | float | 1 | Factor applied to the cores each GPU reports to the scheduler, which is 100 times this value. Accepts decimals, such as 1.5. A value above 1 lets the resourceCoreName requests on one GPU add up to more than 100. Not applied in the mig operating mode. A device-plugin nodeconfig entry can override it per node with devicecorescaling. @schema type: number |
| devices.nvidia.preConfiguredDeviceMemory | int | 0 | Total memory in MiB registered for a GPU whose memory NVML cannot query, such as a unified-memory GPU like the NVIDIA GB10. GPUs that report their memory always use the reported value. 0 skips GPUs whose memory cannot be queried. A device-plugin nodeconfig entry can override it per node with preconfigureddevicememory. |
| devices.nvidia.enableNumaTopology | bool | false | Advertise each GPU's NUMA node on its vGPUs so kubelet's Topology Manager can align CPUs and GPUs. Disabled by default because the single-numa-node Topology Manager policy then changes which Pods kubelet admits. A device-plugin nodeconfig entry can override it per node with enablenumatopology. |
| devices.nvidia.runtimeClassName | string | "" | RuntimeClass set on the device plugin Pods and, by the admission webhook, on GPU Pods that do not set runtimeClassName, commonly nvidia when the NVIDIA container runtime is not the default. The RuntimeClass must exist, either created by createRuntimeClass or already present in the cluster. Empty sets no RuntimeClass. |
| devices.nvidia.createRuntimeClass | bool | false | Create the RuntimeClass named by runtimeClassName, with the nvidia handler, when devicePlugin.enabled is true. Has no effect when runtimeClassName is empty. Leave it false if the RuntimeClass already exists, such as one created by the GPU Operator. |
| devices.nvidia.gpuCorePolicy | string | "default" | How HAMi-core enforces the resourceCoreName limit. The admission webhook passes values other than default as GPU_CORE_UTILIZATION_POLICY. default lets vGPUmonitor turn enforcement on only while it sees other active tasks of the same or higher priority on the GPU. force always enforces it. disable never enforces it. HAMi-core treats any other value as default. Limits of 0 and 100 are never enforced. |
| devices.nvidia.libCudaLogLevel | int | 1 | HAMi-core log level, passed to containers as LIBCUDA_LOG_LEVEL. Errors are always logged. 0 and 1 log errors only, 2 adds warnings and messages, 3 adds info, and 4 adds debug logs. HAMi-core itself defaults to 2 when the variable is unset. Not applied in the mig operating mode. A device-plugin nodeconfig entry can override it per node with libcudaloglevel. |
| devices.nvidia.limitDelivery | string | "env" | How the device plugin passes a container's GPU limits to HAMi-core. file sets the limit environment variables and also mounts the limits read-only at /overrideEnv, so HAMi-core can apply them to processes started without the container's environment, such as from SSH, su -, sudo or cron. env sets the environment variables only. A device-plugin nodeconfig entry can override it per node with limitdelivery, and a Pod with the hami.io/limit-delivery annotation. An unknown value is ignored, and file is used when no valid value is set. Not applied in the mig operating mode. |
| devices.remotegpu.enabled | bool | false | Schedule Pods onto GPUs served over the network by a lupine server. Mark lupine nodes with operatingmode: remote in devicePlugin.nodeConfiguration (peer to hami-core / mig). The device plugin applies hami.io/lupine-server; set that label's value to a custom port only when the server is managed outside the chart (empty value uses defaultPort below). Keep resource names consistent with any device-config.content override. |
| devices.remotegpu.defaultPort | int | 14833 | Port (1-65535) used by the scheduler and the server DaemonSet's container and host ports. A valid port in the hami.io/lupine-server node label overrides the scheduler's connection port. Must match the server's listen port; the chart does not configure the server process. |
| devices.remotegpu.resourceCountName | string | "nvidia.com/remote-gpu" | Resource name for an integer GPU count in a container's requests or limits. A positive count reserves a whole server with at least that many GPUs, including any extra GPUs on that server. Zero requests no GPU allocation. |
| devices.remotegpu.resourceMemoryName | string | "nvidia.com/remote-gpu-memory" | Resource name for optional per-GPU memory in a container's requests or limits, as integer MiB (1024 = 1 GiB). Used with a remote GPU count request. Missing or zero requests set no memory limit. Enforcement requires library injection through devices.remotegpu.libImage. |
| devices.remotegpu.libImage | string | nil | HAMi-core image for the client's library-copy init container, used to enforce positive memory requests. Must provide sh, cp, and /k8s-vgpu/lib/nvidia/libvgpu.so.*. Leave unset or null to use the device plugin image, including its registry and tag overrides. Set to "" to skip library injection and use memory requests only as placement filters. An explicit image reference is used unchanged. |
| devices.remotegpu.clientImage | string | "ghcr.io/lupinemachines/lupine-client:cuda-13.3.1-ubuntu24.04@sha256:fe0e8c283fe29b0c8bfb4bbe2a81fac4fabd3a477fe694f31b5a97a996fb028e" | Client image used to deliver Lupine libraries, libnghttp2, and nvidia-smi for the CUDA/vLLM examples; empty disables delivery. It must contain /opt/lupine/lib, sh, cp, libnghttp2.so.14, and nvidia-smi. Injection fails when nvidia-smi is missing. Match the server version and workload ABI; an empty string disables delivery. global.imageSource=cn prefixes the upstream Lupine client reference with m.daocloud.io/; other repositories are unchanged. |
| devices.remotegpu.sessionImage | string | "" | Image for a relay Pod on each lupine server node, so users can port-forward to the relay. Requires sh and socat; global image overrides do not apply. Empty disables creation of relays without deleting existing ones. |
| devices.remotegpu.sessionImagePullSecrets | list | [] | Names of existing image pull Secrets for the session relay Pods, added after global.imagePullSecrets. The Secrets must exist in the HAMi namespace. |
| devices.remotegpu.server.enabled | bool | true | Deploy a lupine server DaemonSet when devices.remotegpu.enabled is true. Selects only nodes with an empty hami.io/lupine-server label value. A node whose label names its own port is left to a server managed outside the chart. False leaves server deployment to the operator without disabling remote GPU scheduling. |
| devices.remotegpu.server.image.registry | string | "ghcr.io" | Registry of the lupine server container image. global.imageRegistry overrides it. |
| devices.remotegpu.server.image.repository | string | "lupinemachines/lupine-server" | Repository of the lupine server container image. |
| devices.remotegpu.server.image.tag | string | "cuda-13.3.1-ubuntu24.04@sha256:79f023523ed0ea8704b6e3634c5831ed2d4210a0a8c7ed700a3bcdbbf7c9e1a3" | Tag and digest of the lupine server container image; does not inherit global.imageTag. |
| devices.remotegpu.server.image.pullPolicy | string | "IfNotPresent" | Pull policy of the lupine server container image: Always, IfNotPresent, or Never. |
| devices.remotegpu.server.image.pullSecrets | list | [] | Names of existing image pull Secrets for the lupine server Pods, added after global.imagePullSecrets. The Secrets must exist in the HAMi namespace. |
| devices.remotegpu.server.runtimeClassName | string | "" | RuntimeClass for lupine server Pods. Set to a class such as nvidia when the NVIDIA container runtime is not the cluster default. Empty omits the field; independent of devices.nvidia.runtimeClassName. |
| devices.remotegpu.server.tolerations | list | [] | Kubernetes tolerations for lupine server Pods, as a list of objects with key, operator, value, effect, and optional tolerationSeconds. |
| devices.remotegpu.server.securityContext | object | {} | Container securityContext of the lupine server. Set privileged: true on hosts where the NVIDIA container toolkit has accept-nvidia-visible-devices-envvar-when-unprivileged=false, otherwise NVIDIA_VISIBLE_DEVICES is ignored and the server sees no GPU. |
| devices.remotegpu.server.resources | object | {} | Kubernetes resource requests and limits for the lupine server container, for example {requests: {cpu: 100m, memory: 100Mi}, limits: {memory: 1Gi}}. |
| devices.ascend.configs | list | Per-chip capacities and templates; see values.yaml. | Per-chip objects shared with the separately deployed Ascend device plugin as vnpus.configs; overrides replace the whole list. String chipName must match the detected chip name; string commonWord identifies the HAMi device type. resourceName, resourceMemoryName, and resourceCoreName are resource-name strings for Pod device-count, memory, and core requests. Pod counts are non-negative integers. Memory requests and memoryAllocatable, memoryCapacity, and template memory use integer MiB (16384 = 16 GiB), without a Gi suffix. memoryAllocatable is usable memory; memoryCapacity is full-device capacity. memoryFactor is an integer request multiplier, applied only above 1. <br>Pod core requests in hami-core/enpu mode are integer percentages, 0-100 (40 means 40%); ENPU treats 0 or omission as 100%. Configured aiCore/aiCPU values are integer hardware core counts, not percentages or virtual-device slots. templates lists driver slice definitions: a string name, integer memory, and optional integer aiCore/aiCPU. The named template must be supported by the driver; editing these numbers does not create a new driver template. The boolean superPod enables Ascend910C module pairing: non-ENPU requests of 1 become 2; other odd counts are rejected. See single-card soft-slicing workload examples. <br>The Ascend950DT entry is for the 9582 SKU with 96 GiB HBM and requires a compatible separately deployed device plugin. |
| devices.ascend.runtimeClassName | string | "" | RuntimeClass name to set on Pods requesting Ascend resources when their spec.runtimeClassName is unset. Empty leaves the Pod unchanged; an existing Pod value is preserved. |
| devices.ascend.hamiVnpuMode | string | "" | Default VNPU mode: template, hami-core (alias: hamiCore), or enpu. Node capability annotations take priority. An empty or null value uses template, or hami-core when the deprecated devices.ascend.hamiVnpuCore is true. An explicit mode takes precedence over that flag; unknown modes are rejected. |
| devices.ascend.deviceSplitCount | int | 10 | Positive integer virtual-device slots per physical Ascend NPU in hami-core and ENPU modes. Requires an ascend-device-plugin version supporting vnpus.deviceSplitCount (introduced in #144). A positive node vDeviceCount takes precedence; ENPU is capped at 100. Template mode keeps its template calculation. |
| devices.ascend.enpuPolicy | string | "elastic" | Default ENPU policy: fixed-share, elastic, or best-effort; numeric string aliases "1", "2", and "3" are accepted. Empty uses elastic. A Pod's huawei.com/enpu-policy overrides this value; invalid policies are rejected for ENPU workloads. |
| devices.ascend.enabled | bool | false | Enable Ascend support in the HAMi scheduler and add devices.ascend.customresources to the bundled kube-scheduler's extender configuration. Install the Ascend device plugin separately; this chart does not deploy it. |
| devices.ascend.image | string | "" | Not consumed by this chart. Configure the Ascend device plugin image in its separate deployment. |
| devices.ascend.imagePullPolicy | string | "IfNotPresent" | Not consumed by this chart. Configure the Ascend device plugin's image pull policy in its separate deployment. |
| devices.ascend.extraArgs | list | [] | Not consumed by this chart. Configure additional Ascend device plugin arguments in its separate deployment. |
| devices.ascend.nodeSelector | object | {"ascend":"on"} | Not consumed by this chart. Configure the Ascend device plugin Pod's node selector in its separate deployment. |
| devices.ascend.tolerations | list | [] | Not consumed by this chart. Configure the Ascend device plugin Pod's tolerations in its separate deployment. |
| devices.ascend.customresources | list | Ascend device-count, memory, and core resource names; see values.yaml. | Resource-name strings managed by the bundled kube-scheduler's extender when devices.ascend.enabled is true. Include the resource names from devices.ascend.configs; they are not added automatically. This does not register kubelet resources; configure an external kube-scheduler separately. |
| devices.ascend.hamiVnpuCore | bool | false | |
| devices.ascend.hamiCorePolicy | string | "spatial" | |
| devices.ascend.deviceCoreScaling | int | 1 | |
| devices.iluvatar.configs | list | Device definitions for MR-V100, MR-V50, BI-V150, and BI-V100; see values.yaml. | List of Iluvatar device definitions used when devices.iluvatar.enabled is true. Each object contains string fields chipName, commonWord, resourceCountName, resourceMemoryName, and resourceCoreName. commonWord must be unique and match the device plugin's type and annotation names; chipName is not used for matching by the HAMi scheduler. The resource-name strings identify Pod requests: count is a non-negative integer (0 requests no devices); memory is a non-negative integer in 256 MiB units (64 = 16 GiB, without Mi or Gi suffixes). Zero or omitted memory requests use the full device memory. For one GPU, cores are an integer from 0 to 100 (40 means 40%, not 40%); for multiple GPUs, admission sets 100% cores per GPU. See Iluvatar workload examples. Overrides replace the whole list; supply complete objects because omitted fields are not filled from defaults. An empty list removes all Iluvatar device definitions. A non-empty device-config.content or bundled files/device-config.yaml takes precedence over this list. |
| devices.iluvatar.enabled | bool | false | Enable Iluvatar device handling in the HAMi scheduler and include devices.iluvatar.customresources in the bundled kube-scheduler's extender configuration. Requires a separately installed Iluvatar gpu-manager; see Iluvatar setup. |
| devices.iluvatar.customresources | list | Count, core, and memory resource names for MR-V100, MR-V50, BI-V150, and BI-V100; see values.yaml. | Resource-name strings delegated to the HAMi scheduler extender so kube-scheduler skips its own checks for these Iluvatar resources. Used only when both devices.iluvatar.enabled and scheduler.kubeScheduler.enabled are true. Include the resource names from devices.iluvatar.configs; they are not added automatically. Overrides replace the whole list. |
| devices.alibaba.enabled | bool | true | Enable Alibaba PPU device |
| devices.alibaba.ppuCorePolicy | string | "default" | |
| devices.alibaba.customresources | list | ["alibaba.com/ppu","alibaba.com/ppu-memory","alibaba.com/gpu-core"] | Custom resource definitions for Alibaba devices |
| legacyMetrics | bool | false | |
| prometheus.enabled | bool | false | Create scheduler and device-plugin ServiceMonitors when the API is available. The device-plugin ServiceMonitor also requires devicePlugin.enabled. For local helm template, pass --api-versions monitoring.coreos.com/v1/ServiceMonitor. |
| device-config.content | string | "" | Full replacement of device-config.yaml; takes precedence over a bundled file and values. This is a complete document, not a merge; prefer devices.<vendor> fields for partial overrides. Please note that extender resources still use devices.*, keep resource names consistent. Values validation still applies. |
| legacySelectorLabels | bool | false |
<!-- END GENERATED VALUES -->