[源码解析] 深度学习分布式训练框架 horovod (20) --- Elastic Training Operator

2021/7/28 20:06:03

本文主要是介绍[源码解析] 深度学习分布式训练框架 horovod (20) --- Elastic Training Operator,对大家解决编程问题具有一定的参考价值,需要的程序猿们随着小编来一起学习吧!

[源码解析] 深度学习分布式训练框架 horovod (20) --- Elastic Training Operator

目录
  • [源码解析] 深度学习分布式训练框架 horovod (20) --- Elastic Training Operator
    • 0x00 摘要
    • 0x01 背景知识
      • 1.1 已有弹性能力
      • 1.2 mpi-operator 的缺点
    • 0x02 总体架构
      • 2.1 资源创建
      • 2.2 角色
      • 2.3 程序主流程
    • 0x03 入口
      • 3.1 创建
      • 3.2 设置
    • 0x04 TrainingJobReconciler
      • 4.1 Reconcile
      • 4.2 ReconcileJobs
      • 4.3 reconcileResource
      • 4.4 doSteps
      • 4.5 createTrainingJobWorkers
        • 4.5.1 CreateWorkers
          • 4.5.1.1 createWorkers
          • 4.5.1.2 createWorker
          • 4.5.1.3 newService
        • 4.5.2 newWorker
      • 4.8 createLauncher
        • 4.8.1 CreateHostConfigMap
        • 4.8.2 创建pod
          • 4.8.2.1 CreateLauncher
          • 4.8.2.2 newLauncher
    • 0x05 ScaleOut
      • 5.1 思路
      • 5.2 Reconcile
      • 5.3 setScalingOwner
      • 5.4 TrainingJobController
        • 5.4.1 Reconcile
        • 5.4.2 ReconcileJobs
      • 5.5 JobRunning
        • 5.5.1 reconcileJobRunning
        • 5.5.2 setTrainingJobScaler
        • 5.5.3 updateLatestScaler
        • 5.5.4 updateCurrentScaler
        • 5.5.5 updateScalerState
      • 5.6 Scaling
        • 5.6.1 executeScaling
        • 5.6.2 executeScaleOut
        • 5.6.3 executeScaleScript
          • 5.6.3.1 hostfileUpdateScript
          • 5.6.3.2 getHostfileContent
          • 5.6.3.3 executeOnLauncher
          • 5.6.3.4 kubectlOnPod
    • 0x06 ScaleIn
      • 6.1 思路
      • 6.2 Reconcile
      • 6.3 setScalingOwner
      • 6.4 executeScaleIn
      • 6.5 setsSaleInToDelete
      • 6.6 DeleteWorkers
      • 6.7 DeleteWorkerPods
    • 0xEE 个人信息
    • 0xFF 参考

0x00 摘要

Horovod 是一款基于 AllReduce 的分布式训练框架。凭借其对 TensorFlow、PyTorch 等主流深度学习框架的支持,以及通信优化等特点,Horovod 被广泛应用于数据并行的训练中。

本文是 horovod on k8s 的最后一篇,看看 MPI-Operator 可能被如何改进,主要就是根据 Elastic Training Operator 作者 团队的博客内容来学习源码。所以本文以大量源码为主。

本系列其他文章链接如下:

[源码解析] 深度学习分布式训练框架 Horovod (1) --- 基础知识

[源码解析] 深度学习分布式训练框架 horovod (2) --- 从使用者角度切入

[源码解析] 深度学习分布式训练框架 horovod (3) --- Horovodrun背后做了什么

[源码解析] 深度学习分布式训练框架 horovod (4) --- 网络基础 & Driver

[源码解析] 深度学习分布式训练框架 horovod (5) --- 融合框架

[源码解析] 深度学习分布式训练框架 horovod (6) --- 后台线程架构

[源码解析] 深度学习分布式训练框架 horovod (7) --- DistributedOptimizer

[源码解析] 深度学习分布式训练框架 horovod (8) --- on spark

[源码解析] 深度学习分布式训练框架 horovod (9) --- 启动 on spark

[源码解析] 深度学习分布式训练框架 horovod (10) --- run on spark

[源码解析] 深度学习分布式训练框架 horovod (11) --- on spark --- GLOO 方案

[源码解析] 深度学习分布式训练框架 horovod (12) --- 弹性训练总体架构

[源码解析] 深度学习分布式训练框架 horovod (13) --- 弹性训练之 Driver

[源码解析] 深度学习分布式训练框架 horovod (14) --- 弹性训练发现节点 & State

[源码解析] 深度学习分布式训练框架 horovod (15) --- 广播 & 通知

[源码解析] 深度学习分布式训练框架 horovod (16) --- 弹性训练之Worker生命周期

[源码解析] 深度学习分布式训练框架 horovod (17) --- 弹性训练之容错

[源码解析] 深度学习分布式训练框架 horovod (18) --- kubeflow tf-operator

[源码解析] 深度学习分布式训练框架 horovod (17) --- 弹性训练之容错

[源码解析] 深度学习分布式训练框架 horovod (18) --- kubeflow tf-operator

[源码解析] 深度学习分布式训练框架 horovod (19) --- kubeflow MPI-operator

0x01 背景知识

0x01, 0x02 两节均来自于 Elastic Training Operator 团队博客内容,这个博客真得很给力。

1.1 已有弹性能力

Kubernetes 和云计算提供敏捷性和伸缩性,我们可以通过 cluster-AutoScaler 等组件为训练任务设置弹性策略,利用 Kubernetes 的弹性能力,按需创建,减少 GPU 设备空转。

但这种伸缩模式面对训练这种离线任务还是略有不足:

  • 不支持容错,当部分 Worker 由于设备原因失败,整个任务需要停止重来。
  • 训练任务一般时间较长,占用算力大,任务缺少弹性能力。当资源不足时,除非任务终止,无法按需为其他业务腾出资源。
  • 训练任务时间较长,不支持 worker 动态配置, 无法安全地使用抢占实例,发挥云上最大性价比

如何给训练任务赋予弹性能力,是提高性价比的关键路径。近期 horovod 等分布式框架逐渐支持了 Elastic Training,即弹性训练能力。也就是允许一个训练任务在执行的过程中动态的扩容或者缩容训练 worker, 从不会引起训练任务的中断。需要在代码中做少量修改适配,可参考:https://horovod.readthedocs.io/en/stable/elastic_include.html。

1.2 mpi-operator 的缺点

在 mpi-operator 中,参与训练的 Worker 都是作为静态资源设计和维护,支持弹性训练模式后,给任务增加了灵活性,同时也给运维层带来了挑战,例如:

  • 必须通过 horovod 提供的 horovordrun 作为入口,horovod 中 launcher 通过 ssh 登陆 worker,需要打通 launcher 和 worker 之间的登陆隧道。
  • 负责计算弹性的 Elastic Driver 模块通过指定 discover_host 脚本获取最新 worker 拓扑信息,从而拉起或停止 worker 实例。当 worker 变化时,首先要更新 discover_host 脚本的返回值。
  • 在抢占或价格计算等场景中,有时需要指定 worker 缩容,K8s 原生的编排元语 deployment,statefulset 无法满足指定缩容的场景。

针对以上问题,我们设计开发了 et-operator,提供 TrainingJob CRD 描述训练任务, ScaleOut 和 ScaleIn CRD 描述扩容和缩容操作, 通过它们的组合,使我们的训练任务更具有弹性。将这个方案开源,欢迎大家提需求、交流、吐槽。

开源方案地址:https://github.com/AliyunContainerService/et-operator

0x02 总体架构

TrainingJob Controller 主要有以下功能:

  • 维护 TrainingJob 的创建/删除生命周期,以及子资源管理。
  • 执行扩缩容操作。
  • 容错,当 worker 被驱逐,创建新的 worker 加入到训练中。

2.1 资源创建

TrainingJob 子资源创建顺序如下:

  • 创建打通 ssh 所需的密钥对, 创建 secret。
  • 创建 workers,包含 service 和 pod,挂载 secret 公钥。
  • 创建 configmap, 包含 discover_host 脚本 , hostfile 文件。
  • 创建 launcher,挂载 configmap。由于 hostfile 后续会随着拓扑关系修改,所以 hostfile 单独通过 initcontainer 从 configmap 拷贝到单独目录。

TrainingJob 相关资源:

2.png

2.2 角色

TrainingJob CR 的配置分为 Lanucher 和 Worker。在 Launcher 中指定任务的镜像和启动执行, 默认 et-operator 会根据 worker 分配情况,生成一个 hostfile 文件和 discover_host 脚本,discover_host 脚本挂载到 Launcher 的 /etc/edl/discover_hosts.sh 文件, 在入口脚本的 horovodrun 执行中通过 --host-discovery-script 参数指定。在 Worker 设置中指定 worker 的镜像和 GPU 占用 ,并可以通过 maxReplicas / minReplicas 指定 workers 的副本数允许范围。

3.png

2.3 程序主流程

程序主流程图如下:

0x03 入口

其实,学习 ETO 主要就是学习如何扩容和缩容。但是为了学习这个,我们还是需要梳理一下程序逻辑

不熟悉 K8S 的同学顺便也一起看看其 CRD 如何使用。

3.1 创建

入口代码是 main.go/main 函数,从入口可以看出,

  • 生成了 Controller.Manager。
  • 利用这个 Manager,构建了三个 Reconciler :TrainingJobReconciler,ScaleInReconciler,ScaleOutReconciler。
  • 然后启动 Manager;
func main() {
	mgr, err := ctrl.NewManager(ctrl.GetConfigOrDie(), ctrl.Options{
		Scheme:             scheme,
		MetricsBindAddress: metricsAddr,
		LeaderElection:     enableLeaderElection,
		Port:               9443,
	})

	const jobPollInterval = "5s"
  
	if err = controllers.NewReconciler(mgr, parseDurationOrPanic(jobPollInterval)).SetupWithManager(mgr); err != nil {
		os.Exit(1)
	}
	if err = controllers.NewScaleOutReconciler(mgr, parseDurationOrPanic(jobPollInterval)).SetupWithManager(mgr); err != nil {
		os.Exit(1)
	}
	if err = controllers.NewScaleInReconciler(mgr, parseDurationOrPanic(jobPollInterval)).SetupWithManager(mgr); err != nil {
		os.Exit(1)
	}

	if err := mgr.Start(ctrl.SetupSignalHandler()); err != nil {
		os.Exit(1)
	}
}

3.2 设置

这里的配置就是建立了消息的响应函数,具体就是响应哪些 CR。

  • 除了 TrainingJob 外,et-operator 同时支持 ScaleOut 和 ScaleIn 两种 CRD,下发训练任务扩容和缩容操作。

  • 当下发一个 ScaleOut CR,ScaleOutController 触发 Reconcile, 这里工作很简单,根据 ScaleOut CR 中的 Selector 字段,找到 Scaler 对应的 TrainingJob,设置到 CR 的 OwnerReferences 上。

  • TrainingJobController 中监听到属于 TrainingJob 的 ScaleOut CR 有更新, 触发 TrainingJob 的 Reconcile,遍历过滤 TrainingJob 下 OwnerReference 指向的 ScaleIn 和 ScaleOut, 根据创建时间和状态时间决定执行的扩容或者缩容。

  • 执行缩容时,可以通过 ScaleIn CR 中的 spec.toDelete.count 或 spec.toDelete.podNames 字段指定缩容的 worker。通过 count 配置缩容的数量,则通过 index 计算由高到低缩容 Worker。

func (r *ScaleInReconciler) SetupWithManager(mgr ctrl.Manager) error {
	return ctrl.NewControllerManagedBy(mgr).
		For(&kaiv1alpha1.ScaleIn{}).
		Complete(r)
}

func (r *ScaleOutReconciler) SetupWithManager(mgr ctrl.Manager) error {
	return ctrl.NewControllerManagedBy(mgr).
		For(&kaiv1alpha1.ScaleOut{}).
		Complete(r)
}

func (r *TrainingJobReconciler) SetupWithManager(mgr ctrl.Manager) error {
	return ctrl.NewControllerManagedBy(mgr).
		For(&kaiv1alpha1.TrainingJob{}).
		Owns(&kaiv1alpha1.ScaleIn{}).
		Owns(&kaiv1alpha1.ScaleOut{}).
		Owns(&corev1.Pod{}).
		Owns(&corev1.Service{}).
		Owns(&corev1.ConfigMap{}).
		Owns(&corev1.Secret{}).
		// Ignore status-only and metadata-only updates
		//WithEventFilter(predicate.GenerationChangedPredicate{}).
		Complete(r)
}

0x04 TrainingJobReconciler

顺着代码梳理一下,寻找其设计思想精微之处。

4.1 Reconcile

k8s operator 中reconcile方法 的作用就是不断的watch,当资源变化时 就会触发reconcile方法,理论上有多少次的变化就会执行多少次的reconcile方法。

当有消息来的时候,Reconcile 方法会得到调用。

func (r *TrainingJobReconciler) Reconcile(req ctrl.Request) (ctrl.Result, error) {
	// Fetch latest training job instance.
	sharedTrainingJob := &kaiv1alpha1.TrainingJob{}
	err := r.Get(context.Background(), req.NamespacedName, sharedTrainingJob)
	trainingJob := sharedTrainingJob.DeepCopy()
	// Check reconcile is required.
	// No need to do reconcile or job has been deleted.
	r.Scheme.Default(trainingJob)
	return r.ReconcileJobs(trainingJob)
}

4.2 ReconcileJobs

因为消息中状态是 "",所以运行了 initializeJob,并且进行 reconcileResource。

func (r *TrainingJobReconciler) ReconcileJobs(job *kaiv1alpha1.TrainingJob) (result reconcile.Result, err error) {
	oldJobStatus := job.Status.DeepCopy()

	defer func() {
		latestJob := &kaiv1alpha1.TrainingJob{}
		err := r.Get(context.Background(), types.NamespacedName{
			Name:      job.Name,
			Namespace: job.Namespace,
		}, latestJob)
		if err == nil {
			if latestJob.ObjectMeta.ResourceVersion != job.ObjectMeta.ResourceVersion {
				latestJob.Status = job.Status
				job = latestJob
			}
		}
		r.updateObjectStatus(job, oldJobStatus)
	}()

	switch job.Status.Phase {
    case commonv1.JobSucceeded, commonv1.JobFailed:
      err = r.cleanup(job)
    case "", commonv1.JobCreated: // 如果状态为空 或者 JobCreated,则初始化
      r.initializeJob(job)
      err = r.reconcileResource(job)
    case commonv1.JobRunning:
      err = r.reconcileJobRunning(job)
    case commonv1.Scaling:
      err = r.executeScaling(job)
	}

	if err != nil {
		if IsRequeueError(err) {
			return RequeueAfterInterval(r.PollInterval, nil)
		}
		return RequeueAfterInterval(r.PollInterval, err)
	}
	return NoRequeue()
}

4.3 reconcileResource

reconcileResource 其实就是调用 doSteps,调用一个状态机继续初始化。

func (r *TrainingJobReconciler) reconcileResource(job *kaiv1alpha1.TrainingJob) error {
	steps := r.newSteps()
	err := r.doSteps(job, steps)
	return err
}

4.4 doSteps

newSteps 构建了一个简单的状态机,是一个初始化步骤,按照序列执行,doSteps 会根据状态进行不同的分支处理。

有几点需要说明:

  • Created 之后的几个状态,应该是: WorkersCreated ---> WorkersReady ----> LauncherCreated ---> JobRunning
  • 这个是事后状态,即对应 action 完成之后应该达到的状态。
  • 在 for 循环之中,如果当前 Job 已经达到了某个状态,就跳过继续,直到某一个未完状态,就去执行对应的action。所以理论上说,会从 WorkersCreated 逐步执行到 JobRunning。
  • 在某个状态对应的 Action 中,执行完成之后,会设置 Job 为这个 完成状态。

代码如下:

func (r *TrainingJobReconciler) newSteps() []Step {
	return []Step{
		Step{
			JobCondition: commonv1.WorkersCreated,
			Action:       r.createTrainingJobWorkers,
		},
		Step{
			JobCondition: commonv1.WorkersReady,
			Action:       r.waitWorkersRunning,
		},
		Step{
			JobCondition: commonv1.LauncherCreated,
			Action:       r.createLauncher,
		},
		Step{
			JobCondition: commonv1.JobRunning,
			Action:       r.syncLauncherState,
		},
	}
}

func (r *TrainingJobReconciler) doSteps(job *kaiv1alpha1.TrainingJob, steps []Step) error {
	for _, step := range steps {
		if hasCondition(*job.GetJobStatus(), step.JobCondition) {
			continue
		}
		err := step.Action(job)
		break
	}
	return nil
}

所以具体如下:

           Request("")
K8S  +-------------------->  Reconcile
                                 +
                                 |
                                 |
                                 v
          +----------------------+---------------------+
          |                 ReconcileJobs              |
          |                      +                     |
          |                      |                     |
          |        +------------------------------+    |
          |        |             |                |    |
          |        v             v                v    |
          |  "", JobCreated   JobRunning      Scaling  |
          +--------+-----------------------------------+
                   |
                   |
                   v
           reconcileResource
                   +
                   |
                   |
                   v
         +---------+---------------+
         | doSteps                 |
         |                         |
         |                         |
         |     WorkersCreated +---------> createTrainingJobWorkers
         |                         |
         |                         |
         |     WorkersReady  +----------> waitWorkersRunning
         |                         |
         |                         |
         |     LauncherCreated +--------> createLauncher
         |                         |
         |                         |
         |     JobRunning  +------------> syncLauncherState
         |                         |
         +-------------------------+

4.5 createTrainingJobWorkers

在 doSteps 步骤中,先来到 createTrainingJobWorkers 这个Action。这里会设置 Job 状态为 WorkersCreated。

func (r *TrainingJobReconciler) createTrainingJobWorkers(job *kaiv1alpha1.TrainingJob) error {
	if job.GetAttachMode() == kaiv1alpha1.AttachModeSSH {
		if cm, err := r.GetOrCreateSecret(job); cm == nil || err != nil {
			updateStatus(job.GetJobStatus(), common.JobFailed, trainingJobFailedReason, msg)
			return nil
		}
	}

	workers := getJobReplicasWorkers(job)
	job.Status.TargetWorkers = workers
    
    // 创建worker
	if err := r.CreateWorkers(job, workers); err != nil {
		updateStatus(job.GetJobStatus(), common.JobFailed, trainingJobFailedReason, msg)
		return nil
	}
    // 设置新状态
	updateJobConditions(job.GetJobStatus(), common.WorkersCreated, "", msg)
	return nil
}

4.5.1 CreateWorkers

CreateWorkers 会进行创建worker,如本文前面介绍,worker 包含 service 和 pod,所以创建过程具体为:

  • 调用 另一个同名函数CreateWorkers 来间接创建 workerService。

  • 调用 newWorker 去创建 Pod。

func (r *TrainingJobReconciler) CreateWorkers(job *kaiv1alpha1.TrainingJob, workers []string) error {
	return r.createWorkers(job, workers, func(name string, index string) *corev1.Pod {
		worker := newWorker(job, name, index)
		return worker
	})
}

4.5.1.1 createWorkers

这里会循环调用 createWorker 依据配置生成一系列 workers

func (r *TrainingJobReconciler) createWorkers(job *kaiv1alpha1.TrainingJob, workers []string, newPod PodTplGenerator) error {
    // 遍历,创建
	for _, podName := range workers {
		index, err := getWorkerIndex(job.Name, podName)
		if err != nil {
			return err
		}
		_, err = r.createWorker(job, int32(index), newPod)
		if err != nil {
			return err
		}
	}
	return nil
}

4.5.1.2 createWorker

这里会依据参数对 worker Pod 进行判断,如果不存在,则创建 某一个 worker

func (r *TrainingJobReconciler) createWorker(job *kaiv1alpha1.TrainingJob, index int32, workerPodTempl PodTplGenerator) (*corev1.Pod, error) {
	name := getWorkerName(job.Name, int(index))
	indexStr := strconv.Itoa(int(index))
	pod := &corev1.Pod{}
	nsn := types.NamespacedName{
		Name:      name,
		Namespace: job.Namespace,
	}
	err := r.Get(context.Background(), nsn, pod)

	if err != nil {
		// If the worker Pod doesn't exist, we'll create it.
		if errors.IsNotFound(err) {
            // 如果没有pod,这里也可以创建pod
			worker := workerPodTempl(name, indexStr)
			if job.GetAttachMode() == kaiv1alpha1.AttachModeSSH {
				util.MountRsaKey(worker, job.Name)
			}
			if err = r.Create(context.Background(), worker); err != nil {
				return nil, err
			}
		} 
	}

	service := &corev1.Service{}
	err = r.Get(context.Background(), nsn, service)
	if errors.IsNotFound(err) {
        // 调用newService 进行具体创建
		err = r.Create(context.Background(), newService(job, name, indexStr))
	}
	return nil, nil
}

4.5.1.3 newService

这里才来到具体创建service,真是百转千回。

func newService(obj interface{}, name string, index string) *corev1.Service {
	job, _ := obj.(*kaiv1alpha1.TrainingJob)
	labels := GenLabels(job.Name)
	labels[labelTrainingRoleType] = worker
	labels[replicaIndexLabel] = index
	return &corev1.Service{ // 具体创建
		ObjectMeta: metav1.ObjectMeta{
			Name:      name,
			Namespace: job.Namespace,
			Labels:    labels,
			OwnerReferences: []metav1.OwnerReference{
				*metav1.NewControllerRef(job, kaiv1alpha1.SchemeGroupVersionKind),
			},
		},
		Spec: corev1.ServiceSpec{
			ClusterIP: "None",
			Selector:  labels,
			Ports: []corev1.ServicePort{
				{
					Name: "ssh-port",
					Port: 22,
				},
			},
		},
	}
}


4.5.2 newWorker

newWorker 则构建了 Pod,就是比较常见的套路。

func newWorker(obj interface{}, name string, index string) *corev1.Pod {
	job, _ := obj.(*kaiv1alpha1.TrainingJob)
	labels := GenLabels(job.Name)
	labels[labelTrainingRoleType] = worker
	labels[replicaIndexLabel] = index
	podSpec := job.Spec.ETReplicaSpecs.Worker.Template.DeepCopy()

	// keep the labels which are set in PodTemplate
	if len(podSpec.Labels) == 0 {
		podSpec.Labels = make(map[string]string)
	}
	for key, value := range labels {
		podSpec.Labels[key] = value
	}

	// RestartPolicy=Never
	setRestartPolicy(podSpec)

	container := podSpec.Spec.Containers[0]

	// if we want to use ssh, will start sshd service firstly.
	if len(container.Command) == 0 {
		if job.GetAttachMode() == kaiv1alpha1.AttachModeSSH {
			container.Command = []string{"sh", "-c", "/usr/sbin/sshd  && sleep 365d"}
		} else {
			container.Command = []string{"sh", "-c", "sleep 365d"}
		}
	}
	podSpec.Spec.Containers[0] = container

    // 创建了pod
	return &corev1.Pod{
		ObjectMeta: metav1.ObjectMeta{
			Name:        name,
			Namespace:   job.Namespace,
			Labels:      podSpec.Labels,
			Annotations: podSpec.Annotations,
			OwnerReferences: []metav1.OwnerReference{
				*metav1.NewControllerRef(job, kaiv1alpha1.SchemeGroupVersionKind),
			},
		},
		Spec: podSpec.Spec,
	}
}

逻辑如下:

           Request("")
K8S  +-------------------->  Reconcile
                                 +
                                 |
                                 |
                                 v
          +----------------------+---------------------+
          |                 ReconcileJobs              |
          |                      +                     |
          |                      |                     |
          |        +------------------------------+    |
          |        |             |                |    |
          |        v             v                v    |
          |  "", JobCreated   JobRunning      Scaling  |
          +--------+-----------------------------------+
                   |
                   |
                   v
           reconcileResource
                   +
                   |
                   |
                   v
         +---------+---------------+
         | doSteps                 |                                           +----> createWorkers +----> createWorker +----> newService
         |                         |                                           |
         |                         |                                           +
         |     WorkersCreated +---------> createTrainingJobWorkers +-----> CreateWorkers  +------->  newWorker +------> WorkersCreated
         |                         |
         |                         |
         |     WorkersReady  +----------> waitWorkersRunning
         |                         |
         |                         |
         |     LauncherCreated +--------> createLauncher
         |                         |
         |                         |
         |     JobRunning  +------------> syncLauncherState
         |                         |
         +-------------------------+

手机如下:

4.8 createLauncher

建立完 worker 之后,就开始建立 Launcher。所以继续执行 createLauncher。

func (r *TrainingJobReconciler) createLauncher(job *kaiv1alpha1.TrainingJob) error {
	if _, err := r.GetOrCreateLauncherServiceAccount(job); err != nil {
		updateStatus(job.GetJobStatus(), commonv1.JobFailed, trainingJobFailedReason, msg)
		return nil
	}
	if _, err := r.GetOrCreateLauncherRole(job, 0); err != nil {
		updateStatus(job.GetJobStatus(), commonv1.JobFailed, trainingJobFailedReason, msg)
		return nil
	}
	if _, err := r.GetLauncherRoleBinding(job); err != nil {
		updateStatus(job.GetJobStatus(), commonv1.JobFailed, trainingJobFailedReason, msg)
		return nil
	}

	if cm, err := r.CreateHostConfigMap(job); cm == nil || err != nil {
		updateStatus(job.GetJobStatus(), commonv1.JobFailed, trainingJobFailedReason, msg)
		return nil
	}

	launcher, err := r.GetLauncherJob(job)

	if launcher == nil {
		if _, err := r.CreateLauncher(job); err != nil {
			updateStatus(job.GetJobStatus(), commonv1.JobFailed, trainingJobFailedReason, msg)
			return nil
		}
	}

	updateJobConditions(job.GetJobStatus(), commonv1.LauncherCreated, "", msg)
	return nil
}

我们取两个重点步骤。

4.8.1 CreateHostConfigMap

这里获取关于host的配置。

func (r *TrainingJobReconciler) CreateHostConfigMap(job *kaiv1alpha1.TrainingJob) (*corev1.ConfigMap, error) {
	return r.createConfigMap(job, newHostfileConfigMap)
}

func (r *TrainingJobReconciler) createConfigMap(job *kaiv1alpha1.TrainingJob, newCm func(job *kaiv1alpha1.TrainingJob) *corev1.ConfigMap) (*corev1.ConfigMap, error) {
	cm := &corev1.ConfigMap{}
	name := ctrl.Request{}
	name.NamespacedName.Namespace = job.GetNamespace()
	name.NamespacedName.Name = job.GetName() + configSuffix
	err := r.Get(context.Background(), name.NamespacedName, cm)
	if errors.IsNotFound(err) {
		if err = r.Create(context.Background(), newCm(job)); err != nil {
			return cm, err
		}
	}
	return cm, nil
}

4.8.2 创建pod

4.8.2.1 CreateLauncher

这里进行pod的创建

func (r *TrainingJobReconciler) CreateLauncher(obj interface{}) (*corev1.Pod, error) {
	job, ok := obj.(*kaiv1alpha1.TrainingJob)
	launcher := newLauncher(job) // 创建pod
	if job.GetAttachMode() == kaiv1alpha1.AttachModeSSH {
		util.MountRsaKey(launcher, job.Name)
	}
	err := r.Create(context.Background(), launcher)
	return launcher, nil
}

4.8.2.2 newLauncher

这里就是具体构建 Pod。

func newLauncher(obj interface{}) *corev1.Pod {
	job, _ := obj.(*kaiv1alpha1.TrainingJob)
	launcherName := job.Name + launcherSuffix
	labels := GenLabels(job.Name)
	labels[labelTrainingRoleType] = launcher
	podSpec := job.Spec.ETReplicaSpecs.Launcher.Template.DeepCopy()
	// copy the labels and annotations to pod from PodTemplate
	if len(podSpec.Labels) == 0 {
		podSpec.Labels = make(map[string]string)
	}
	for key, value := range labels {
		podSpec.Labels[key] = value
	}
	podSpec.Spec.InitContainers = append(podSpec.Spec.InitContainers, initContainer(job))
	//podSpec.Spec.InitContainers = append(podSpec.Spec.InitContainers, kubedeliveryContainer())

	container := podSpec.Spec.Containers[0]
	container.VolumeMounts = append(container.VolumeMounts,
		corev1.VolumeMount{
			Name:      hostfileVolumeName,
			MountPath: hostfileMountPath,
		},
		corev1.VolumeMount{
			Name:      configVolumeName,
			MountPath: configMountPath,
		},
		corev1.VolumeMount{
			Name:      kubectlVolumeName,
			MountPath: kubectlMountPath,
		})

	if job.GetAttachMode() == kaiv1alpha1.AttachModeKubexec {
		container.Env = append(container.Env, corev1.EnvVar{
			Name:  "OMPI_MCA_plm_rsh_agent",
			Value: getKubexecPath(),
		})
	}
	podSpec.Spec.Containers[0] = container
	podSpec.Spec.ServiceAccountName = launcherName

	setRestartPolicy(podSpec)
	hostfileMode := int32(0444)
	scriptMode := int32(0555)

	podSpec.Spec.Volumes = append(podSpec.Spec.Volumes,
		corev1.Volume{
			Name: hostfileVolumeName,
			VolumeSource: corev1.VolumeSource{
				EmptyDir: &corev1.EmptyDirVolumeSource{},
			},
		},
		corev1.Volume{
			Name: kubectlVolumeName,
			VolumeSource: corev1.VolumeSource{
				EmptyDir: &corev1.EmptyDirVolumeSource{},
			},
		},
		corev1.Volume{
			Name: configVolumeName,
			VolumeSource: corev1.VolumeSource{
				ConfigMap: &corev1.ConfigMapVolumeSource{
					LocalObjectReference: corev1.LocalObjectReference{
						Name: job.Name + configSuffix,
					},
					Items: []corev1.KeyToPath{
						{
							Key:  hostfileName,
							Path: hostfileName,
							Mode: &hostfileMode,
						},
						{
							Key:  discoverHostName,
							Path: discoverHostName,
							Mode: &hostfileMode,
						},
						{
							Key:  kubexeclFileName,
							Path: kubexeclFileName,
							Mode: &scriptMode,
						},
					},
				},
			},
		})
	return &corev1.Pod{
		ObjectMeta: metav1.ObjectMeta{
			Name:        launcherName,
			Namespace:   job.Namespace,
			Labels:      podSpec.Labels,
			Annotations: podSpec.Annotations,
			OwnerReferences: []metav1.OwnerReference{
				*metav1.NewControllerRef(job, kaiv1alpha1.SchemeGroupVersionKind),
			},
		},
		Spec: podSpec.Spec,
	}
}

至此,一个新的训练job被运行起来,其逻辑拓展如下:

           Request("")
K8S  --------------------->  Reconcile
                                 +
                                 |
                                 |
                                 v
          +----------------------+---------------------+
          |                 ReconcileJobs              |
          |                      +                     |
          |                      |                     |
          |        +------------------------------+    |
          |        |             |                |    |
          |        v             v                v    |
          |  "", JobCreated   JobRunning      Scaling  |
          +--------+-----------------------------------+
                   |
                   |
                   v
           reconcileResource
                   +
                   |
                   |
                   v
         +---------+---------------+
         | doSteps                 |                                           +----> createWorkers +----> createWorker +----> newService
         |                         |                                           |
         |                         |                                           |
         |     WorkersCreated +---------> createTrainingJobWorkers +-----> CreateWorkers  +------->  newWorker +------> WorkersCreated
         |                         |
         |                         |
         |     WorkersReady  +----------> waitWorkersRunning
         |                         |
         |                         |
         |     LauncherCreated +--------> createLauncher+----> CreateHostConfigMap +-----> CreateLauncher  +------>  newLauncher
         |                         |
         |                         |
         |     JobRunning  +------------> syncLauncherState
         |                         |
         +-------------------------+

手机如下:

完成了新job的创建,我们看看本文的关键技术点,scaleOut 和 scaleIn。

0x05 ScaleOut

5.1 思路

ScaleOut 任务 CR如下:

4.png

当下发一个 ScaleOut CR,ScaleOutController 触发 Reconcile, 这里工作很简单,根据 ScaleOut CR 中的 Selector 字段,找到 Scaler 对应的 TrainingJob,设置到 CR 的 OwnerReferences 上。

以一个 ScaleOut 操作举例:

- apiVersion: kai.alibabacloud.com/v1alpha1
  kind: ScaleOut
  metadata:
    creationTimestamp: "2020-11-04T13:54:26Z
    name: scaleout-ptfnk
    namespace: default
    ownerReferences:
    - apiVersion: kai.alibabacloud.com/v1alpha1
      blockOwnerDeletion: true
      controller: true
      kind: TrainingJob
      name: elastic-training // 指向扩容对象TrainingJob
      uid: 075b9c4a-22f9-40ce-83c7-656b329a2b9e
  spec:
  selector:
    name: elastic-training
  toAdd:
    count: 2

5.2 Reconcile

当下发一个 ScaleOut CR,ScaleOutController 触发 Reconcile。主要就是调用 setScalingOwner。

func (r *ScaleOutReconciler) Reconcile(req ctrl.Request) (ctrl.Result, error) {
	scaleOut, err := getScaleOut(req.NamespacedName, r.Client)
	if err != nil {
		// Error reading the object - requeue the request.
		return RequeueImmediately()
	}
	if scaleOut == nil || scaleOut.DeletionTimestamp != nil {
		return NoRequeue()
	}

	if isScaleFinished(*scaleOut.GetJobStatus()) {
		return NoRequeue()
	}

  return setScalingOwner(r, scaleOut, r.PollInterval)
}

5.3 setScalingOwner

setScalingOwner 是关键之一。

这里主要是处理当 ScaleOut CR 没有设置 OwnerReferences 的情况,就设置一个。

逻辑是 根据 ScaleOut CR 中的 Selector 字段,找到 Scaler 对应的 TrainingJob,设置到 CR 的 OwnerReferences 上。

func setScalingOwner(r client.Client, scaler Scaler, pollInterval time.Duration) (ctrl.Result, error) {
	ownerRefs := scaler.GetOwnerReferences()
	if len(ownerRefs) == 0 {
		trainingJob := &kaiv1alpha1.TrainingJob{}
		nsn := types.NamespacedName{}
		nsn.Namespace = scaler.GetNamespace()
		nsn.Name = scaler.GetSelector().Name
		err := r.Get(context.Background(), nsn, trainingJob)
		gvk := kaiv1alpha1.SchemeGroupVersionKind
		ownerRefs = append(ownerRefs, *metav1.NewControllerRef(trainingJob, schema.GroupVersionKind{Group: gvk.Group, Version: gvk.Version, Kind: gvk.Kind}))
		scaler.SetOwnerReferences(ownerRefs)

		initializeJobStatus(scaler.GetJobStatus())
		updateJobConditions(scaler.GetJobStatus(), v1.JobCreated, "", msg)
		err = r.Status().Update(context.Background(), scaler)
		err = r.Update(context.Background(), scaler)
	}
	return NoRequeue()
}

// RequeueAfterInterval requeues after a duration when duration > 0 is specified.
func RequeueAfterInterval(interval time.Duration, err error) (ctrl.Result, error) {
	return ctrl.Result{RequeueAfter: interval}, err
}

5.4 TrainingJobController

TrainingJobController 中监听到属于 TrainingJob 的 ScaleOut CR 有更新, 触发 TrainingJob 的 Reconcile,遍历过滤 TrainingJob 下 OwnerReference 指向的 ScaleIn 和 ScaleOut, 根据创建时间和状态时间决定执行的扩容或者缩容

5.4.1 Reconcile

func (r *TrainingJobReconciler) Reconcile(req ctrl.Request) (ctrl.Result, error) {

	rlog := r.Log.WithValues("trainingjob", req.NamespacedName)
	// Fetch latest training job instance.
	sharedTrainingJob := &kaiv1alpha1.TrainingJob{}
	err := r.Get(context.Background(), req.NamespacedName, sharedTrainingJob)

	trainingJob := sharedTrainingJob.DeepCopy()
	// Check reconcile is required.
	// No need to do reconcile or job has been deleted.

	r.Scheme.Default(trainingJob)

	return r.ReconcileJobs(trainingJob)
}

5.4.2 ReconcileJobs

func (r *TrainingJobReconciler) ReconcileJobs(job *kaiv1alpha1.TrainingJob) (result reconcile.Result, err error) {
	oldJobStatus := job.Status.DeepCopy()

	logger.Infof("jobName: %v, phase %s", job.Name, job.Status.Phase)

	defer func() {
		latestJob := &kaiv1alpha1.TrainingJob{}
		err := r.Get(context.Background(), types.NamespacedName{
			Name:      job.Name,
			Namespace: job.Namespace,
		}, latestJob)
		if err == nil {
			if latestJob.ObjectMeta.ResourceVersion != job.ObjectMeta.ResourceVersion {
				latestJob.Status = job.Status
				job = latestJob
			}
		}
		r.updateObjectStatus(job, oldJobStatus)
	}()

	switch job.Status.Phase {
	case commonv1.JobSucceeded, commonv1.JobFailed:
		err = r.cleanup(job)
	case "", commonv1.JobCreated:
		r.initializeJob(job)
		err = r.reconcileResource(job)
	case commonv1.JobRunning:
		err = r.reconcileJobRunning(job)
	case commonv1.Scaling:
		err = r.executeScaling(job)
	default:
		logger.Warnf("job %s unknown status %s", job.Name, job.Status.Phase)
	}

	if err != nil {
		if IsRequeueError(err) {
			return RequeueAfterInterval(r.PollInterval, nil)
		}
		return RequeueAfterInterval(r.PollInterval, err)
	}
	return NoRequeue()
}

以下根据当前 job 状态不同,就有两条线,先是 JobRunning ,然后是 Scaling,最后恢复成 JobRunning。

我们一一分析。

5.5 JobRunning

首先是来到 JobRunning 状态,我们依次看看如何处理。

5.5.1 reconcileJobRunning

func (r *TrainingJobReconciler) reconcileJobRunning(job *kaiv1alpha1.TrainingJob) error {
	if err := r.syncLauncherState(job); err != nil {
		return err
	}
	if err := r.syncWorkersState(job); err != nil {
		return err
	}

	if job.Status.Phase == commonv1.JobRunning {
		return r.setTrainingJobScaler(job) // 既然是JobRunning状态,就可以开始进行设置scaler
	}

	return nil
}

5.5.2 setTrainingJobScaler

首先,通过 availableScaleOutList 或者 availableScaleInList ,然后进行update。

func (r *TrainingJobReconciler) setTrainingJobScaler(job *kaiv1alpha1.TrainingJob) error {
	scaleOut, err := r.availableScaleOutList(job) // 找到scaleout list

	scaleIn, err := r.availableScaleInList(job) // 找到scaleIn list

	scalerList := append(scaleOut, scaleIn...) // 合并

	// Select the latest scaling job
	r.updateLatestScaler(job, scalerList) // 开始设置
	return nil
}

5.5.3 updateLatestScaler

依据创建时间和状态时间,找到最后一个Scaler。

func (r *TrainingJobReconciler) updateLatestScaler(job *kaiv1alpha1.TrainingJob, scalers []Scaler) error {
	var latestScaler Scaler
	if len(scalers) == 0 {
		return nil
	}
	for i, _ := range scalers {
		scalerItem := scalers[i]
        // 依据创建时间和状态时间,找到最后一个Scaler
		if latestScaler == nil || latestScaler.GetCreationTimestamp().Time.Before(scalerItem.GetCreationTimestamp().Time) {
			latestScaler = scalerItem
		}
	}
	return r.updateCurrentScaler(job, latestScaler)
}

5.5.4 updateCurrentScaler

对找到的scaler进行设置。

func (r *TrainingJobReconciler) updateCurrentScaler(job *kaiv1alpha1.TrainingJob, scaleItem Scaler) error {
	job.Status.CurrentScaler = scaleItem.GetFullName()
	msg := fmt.Sprintf("trainingJobob(%s/%s) execute %s", job.Namespace, job.Name, scaleItem.GetFullName())
    
    // 设置状态
	r.updateScalerState(scaleItem, job, newCondition(common.Scaling, scalingStartReason, msg))

	if err := r.updateObjectStatus(scaleItem, nil); err != nil {
		return err
	}
	return nil
}

5.5.5 updateScalerState

这时候会设置 common.Scaling。所以下次运行,会到 Scaling 分支。

func (r *TrainingJobReconciler) updateScalerState(scaleObj Scaler, trainingJob *kaiv1alpha1.TrainingJob, condition common.JobCondition) error {
    
	jobPhase := common.Scaling // 设置 common.Scaling。所以下次运行,会到 Scaling 分支
	currentJob := scaleObj.GetFullName()
	if condition.Type == common.ScaleSucceeded || condition.Type == common.ScaleFailed {
		jobPhase = common.JobRunning
		currentJob = ""
	}

	setCondition(trainingJob.GetJobStatus(), condition)
	updateStatusPhase(trainingJob.GetJobStatus(), jobPhase)
	updateTrainingJobCurrentScaler(trainingJob.GetJobStatus(), currentJob)

	setCondition(scaleObj.GetJobStatus(), condition)
	updateStatusPhase(scaleObj.GetJobStatus(), condition.Type)

	return nil
}

逻辑如下:

           1 Request("")
  K8S  +-------------------->  Reconcile  <------------------+
           2 ScaleOut CR           +                         |
  K8S  +-------------------->      |                         |
                                   |                         |
                                   v                         |
            +----------------------+---------------------+   |
            |                 ReconcileJobs              |   |
            |                      +                     |   |
            |                      |                     |   |
            |        +------------------------------+    |   |
            |     1  |             | 2            3 |    |   |
            |        v             v                v    |   |
            |  "", JobCreated   JobRunning      Scaling  |   |
            +--------+-------------+---------------------+   |
                     |             |                         |
                  1  |             | 2                       |
                     v             v                         |
             reconcileResource   reconcileJobRunning         |
                     +             +                         |
                  1  |             | 2                       |
                     |             |                         |
                     v             v                         |
+--------------------+----+      setTrainingJobScaler        |
| doSteps                 |        +                         |
|                         |        | 2                       |
|                         |        |                         |
|     WorkersCreated      |        v                         |
|                         |      updateScalerState           |
|                         |        +                         |
|     WorkersReady        |        |                         |
|                         |        | 2                       |
|                         |        v                         |
|     LauncherCreated     |      common.Scaling              |
|                         |        +                         |
|                         |        |                         |
|     JobRunning          |        | 2                       |
|                         |        |                         |
+-------------------------+        +-------------------------+


5.6 Scaling

5.6.1 executeScaling

依据 scale 的类型不同,进行不同扩展。

func (r *TrainingJobReconciler) executeScaling(job *kaiv1alpha1.TrainingJob) error {
	if err := r.syncLauncherState(job); err != nil {
		return err
	}

	if job.Status.CurrentScaler == "" {
		updateStatusPhase(job.GetJobStatus(), common.JobRunning)
		return nil
	}

	if isFinished(*job.GetJobStatus()) {
		return nil
	}

	scalerType, scalerName := getScalerName(job.Status.CurrentScaler)
    // 根据 in 还是 out 进行不同的处理
	if scalerType == "ScaleIn" {
		scaleIn, err := getScaleIn(scalerName, r)

		if scaleIn == nil || isScaleFinished(*scaleIn.GetJobStatus()) {
			finishTrainingScaler(job.GetJobStatus())
			return nil
		}

		oldStatus := scaleIn.Status.DeepCopy()
		defer r.updateObjectStatus(scaleIn, oldStatus)

        // 执行具体缩容操作
		if err = r.executeScaleIn(job, scaleIn); err != nil {
			return err
		}
	} else if scalerType == "ScaleOut" {
		scaleOut, err := getScaleOut(scalerName, r)

		if scaleOut == nil || isScaleFinished(*scaleOut.GetJobStatus()) {
			finishTrainingScaler(job.GetJobStatus())
			return nil
		}

		oldStatus := scaleOut.Status.DeepCopy()
		defer r.updateObjectStatus(scaleOut, oldStatus)

        // 执行具体扩容操作
		if err = r.executeScaleOut(job, scaleOut); err != nil {
		}
	}
	return nil
}

5.6.2 executeScaleOut

进行扩展。

  • 使用 setScaleOutWorkers 对 scaleOut.Status.AddPods 进行添加新 pods。
  • 使用 workersAfterScaler 得到 最终的 worker。
  • 使用 executeScaleScript 进行scale 操作。
func (r *TrainingJobReconciler) executeScaleOut(job *kaiv1alpha1.TrainingJob, scaleOut *kaiv1alpha1.ScaleOut) error {

  initializeJobStatus(scaleOut.GetJobStatus())

	if err := r.validateScaleOut(scaleOut); err != nil {
		r.updateScalerFailed(scaleOut, job, err.Error())
		return err
	}

	if err := r.setScaleOutWorkers(job, scaleOut); err != nil {
		return err
	}

	err := r.ScaleOutWorkers(job, scaleOut)
	if err != nil {
		msg := fmt.Sprintf("%s create scaleout workers failed, error: %v", scaleOut.GetFullName(), err)
		r.ScaleOutFailed(job, scaleOut, msg)
		return err
	}

	scaleOutWorkers, err := r.getScalerOutWorkers(job, scaleOut)

	workerStatuses, _ := r.workerReplicasStatus(scaleOut.GetJobStatus(), scaleOutWorkers)

	if workerStatuses.Active < *scaleOut.Spec.ToAdd.Count {
		if IsScaleOutTimeout(scaleOut) {
			msg := fmt.Sprintf("scaleout job %s execution timeout", scaleOut.GetFullName())
			r.ScaleOutFailed(job, scaleOut, msg)
		}
		return NewRequeueError(fmt.Errorf("wait for workers running"))
	}

	hostWorkers := r.workersAfterScaler(job.Status.CurrentWorkers, scaleOut)

	// execute scalein script
    // 执行scale脚本
	if err := r.executeScaleScript(job, scaleOut, hostWorkers); err != nil {
		msg := fmt.Sprintf("%s execute script failed, error: %v", scaleOut.GetFullName(), err)
		r.ScaleOutFailed(job, scaleOut, msg)
		return err
	} else {
		job.Status.TargetWorkers = r.workersAfterScaler(job.Status.TargetWorkers, scaleOut)
		r.updateScalerSuccessd(scaleOut, job)
	}

	return nil
}

5.6.3 executeScaleScript

这时候调用 hostfileUpdateScript,更新 host file;

最终调用 executeOnLauncher执行脚本。

func (r *TrainingJobReconciler) executeScaleScript(trainingJob *kaiv1alpha1.TrainingJob, scaler Scaler, workers []string) error {
	if isScriptExecuted(*scaler.GetJobStatus()) {
		return nil
	}
	msg := fmt.Sprintf("trainingjob(%s/%s): execute script on launcher for %s", trainingJob.Namespace, trainingJob.Name, scaler.GetFullName())

	slots := getSlots(trainingJob)
	scriptSpec := scaler.GetScriptSpec()

	var script string
    // 得到脚本
	if scriptSpec.Script != "" {
		script = scalerScript(scriptSpec.GetTimeout(), scriptSpec.Env, scriptSpec.Script, scaler.GetPodNames(), slots)
	} else {
		hostfilePath := getHostfilePath(trainingJob)
		script = hostfileUpdateScript(hostfilePath, workers, slots)
	}

    // 执行脚本
	_, _, err := r.executeOnLauncher(trainingJob, script)

	updateJobConditions(scaler.GetJobStatus(), common.ScriptExecuted, "", msg)
	return nil
}

5.6.3.1 hostfileUpdateScript

得到最终的脚本string。

func hostfileUpdateScript(hostfile string, workers []string, slot int) string {
	return fmt.Sprintf(
		`echo '%s' > %s`, getHostfileContent(workers, slot), hostfile)
}

5.6.3.2 getHostfileContent

获取host file内容

func getHostfileContent(workers []string, slot int) string {
	var buffer bytes.Buffer
	for _, worker := range workers {
		buffer.WriteString(fmt.Sprintf("%s:%d\n", worker, slot))
	}
	return buffer.String()
}

5.6.3.3 executeOnLauncher

在pod上执行

func (r *TrainingJobReconciler) executeOnLauncher(trainingJob *kaiv1alpha1.TrainingJob, script string) (string, string, error) {
	var err error
	var launcherPod *corev1.Pod
	if launcherPod, err = r.GetLauncherJob(trainingJob); err != nil {
	}

	if launcherPod != nil {
		stdOut, stdErr, err := kubectlOnPod(launcherPod, script)
		return stdOut, stdErr, nil
	}
	return "", "", nil
}


5.6.3.4 kubectlOnPod

拉动 worker。

func kubectlOnPod(pod *corev1.Pod, cmd string) (string, string, error) {
	cmds := []string{
		"/bin/sh",
		"-c",
		cmd,
	}
	stdout, stderr, err := util.ExecCommandInContainerWithFullOutput(pod.Name, pod.Spec.Containers[0].Name, pod.Namespace, cmds)
	if err != nil {
		return stdout, stderr, err
	}
	return stdout, stderr, nil
}

逻辑如下:

           1 Request("")
  K8S  +-------------------->  Reconcile  <------------------+
           2 ScaleOut CR           +                         |
  K8S  +-------------------->      |                         |
                                   |                         |
                                   v                         |
            +----------------------+---------------------+   |
            |                 ReconcileJobs              |   |
            |                      +                     |   |
            |                      |                     |   |
            |        +------------------------------+    |   |
            |     1  |             | 2            3 |    |   |
            |        v             v                v    |   |   3
            |  "", JobCreated   JobRunning      Scaling +----------->  executeScaling
            +--------+-------------+---------------------+   |              +
                     |             |                         |              |
                  1  |             | 2                       |              | 3
                     v             v                         |              v
             reconcileResource   reconcileJobRunning         |        executeScaleOut
                     +             +                         |              +
                  1  |             | 2                       |              |
                     |             |                         |              | 3
                     v             v                         |              v
+--------------------+----+      setTrainingJobScaler        |      executeScaleScript
| doSteps                 |        +                         |              +
|                         |        | 2                       |              |
|                         |        |                         |              | 3
|     WorkersCreated      |        v                         |              v
|                         |      updateScalerState           |     hostfileUpdateScript
|                         |        +                         |              +
|     WorkersReady        |        |                         |              | 3
|                         |        | 2                       |              |
|                         |        v                         |              v
|     LauncherCreated     |      common.Scaling              |       executeOnLauncher
|                         |        +                         |              +
|                         |        |                         |              |
|     JobRunning          |        | 2                       |              | 3
|                         |        |                         |              v
+-------------------------+        +-------------------------+         kubectlOnPod


0x06 ScaleIn

6.1 思路

ScaleIn 任务 CR如下:

5.png

执行缩容时,可以通过 ScaleIn CR 中的 spec.toDelete.count 或 spec.toDelete.podNames 字段指定缩容的 worker。

通过 count 配置缩容的数量,则通过 index 计算由高到低缩容 Worker。

apiVersion: kai.alibabacloud.com/v1alpha1
kind: ScaleIn
metadata:
  name: scalein-workers
spec:
  selector:
    name: elastic-training
  toDelete:
    count: 1

如果想要缩容特定的 Worker,可以配置 podNames:

apiVersion: kai.alibabacloud.com/v1alpha1
kind: ScaleIn
metadata:
  name: scalein-workers
spec:
  selector:
    name: elastic-training
  toDelete:
    podNames:
    - elastic-training-worker-1

运行一个缩容示例,指定数量缩容 1 个 worker:

kubectl create -f examples/scale_in_count.yaml

6.2 Reconcile

当下发一个 scaleInCR,Controller 触发 Reconcile。主要就是调用 setScalingOwner。

func (r *ScaleInReconciler) Reconcile(req ctrl.Request) (ctrl.Result, error) {
	//silog := r.Log.WithValues("scalein", req.NamespacedName)
	scaleIn, err := getScaleIn(req.NamespacedName, r.Client)

	if isScaleFinished(*scaleIn.GetJobStatus()) {
		return NoRequeue()
	}

    // 以上基本都是各种校验
	return setScalingOwner(r, scaleIn, r.PollInterval)
}

6.3 setScalingOwner

setScalingOwner 是关键之一。

这里主要是处理当 ScaleIn CR 没有设置 OwnerReferences 的情况,就设置一个。

逻辑是 根据 ScaleIn CR 中的 Selector 字段,找到 Scaler 对应的 TrainingJob,设置到 CR 的 OwnerReferences 上。

下面移除各种错误检查代码。

func setScalingOwner(r client.Client, scaler Scaler, pollInterval time.Duration) (ctrl.Result, error) {
	ownerRefs := scaler.GetOwnerReferences()
	if len(ownerRefs) == 0 {
		trainingJob := &kaiv1alpha1.TrainingJob{}
		nsn := types.NamespacedName{}
		nsn.Namespace = scaler.GetNamespace()
		nsn.Name = scaler.GetSelector().Name
		err := r.Get(context.Background(), nsn, trainingJob)

		gvk := kaiv1alpha1.SchemeGroupVersionKind
		ownerRefs = append(ownerRefs, *metav1.NewControllerRef(trainingJob, schema.GroupVersionKind{Group: gvk.Group, Version: gvk.Version, Kind: gvk.Kind}))
		scaler.SetOwnerReferences(ownerRefs)

		initializeJobStatus(scaler.GetJobStatus())
		updateJobConditions(scaler.GetJobStatus(), v1.JobCreated, "", msg)
		err = r.Status().Update(context.Background(), scaler)
		err = r.Update(context.Background(), scaler)
	}
	return NoRequeue()
}

6.4 executeScaleIn

JobRunning 状态处理与 ScaleOut类似,所以略过,直接看处理executeScaleIn。

执行缩容时,可以通过 ScaleIn CR 中的 spec.toDelete.count 或 spec.toDelete.podNames 字段指定缩容的 worker。

通过 count 配置缩容的数量,则通过 index 计算由高到低缩容 Worker。

具体结合代码就是:

setsSaleInToDelete 指定哪些需要删除;

executeScaleScript 执行脚本;

DeleteWorkers 删除 worker;

func (r *TrainingJobReconciler) executeScaleIn(job *kaiv1alpha1.TrainingJob, scaleIn *kaiv1alpha1.ScaleIn) error {
	if scaleIn.DeletionTimestamp != nil || isScaleFinished(*scaleIn.GetJobStatus()) {
		logger.Info("reconcile cancelled, scalein does not need to do reconcile or has been deleted")
		return nil
	}

	initializeJobStatus(scaleIn.GetJobStatus())

	//TODO: Validate the scalein count for minSize
	err := r.setsSaleInToDelete(job, scaleIn)

	currentWorkers := r.workersAfterScaler(job.Status.CurrentWorkers, scaleIn)

	// execute scalein script
	if err := r.executeScaleScript(job, scaleIn, currentWorkers); err != nil {
		msg := fmt.Sprintf("%s execute script failed, error: %v", scaleIn.GetFullName(), err)
		r.updateScalerFailed(scaleIn, job, msg)
		return nil
	}

	toDeleteWorkers := scaleIn.GetPodNames()
	remainWorkers := false
	if scaleIn.Spec.Script == "" {
		if shutdownWorkers, err := r.checkWorkerShutdown(job, toDeleteWorkers); err != nil {
			return err
		} else {
			if len(toDeleteWorkers) != len(shutdownWorkers) {
				remainWorkers = true
				toDeleteWorkers = shutdownWorkers
			}
		}
	}
	if err := r.DeleteWorkers(job, toDeleteWorkers); err != nil {
		msg := fmt.Sprintf("%s delete resource failed, error: %v", scaleIn.GetFullName(), err)
		r.updateScalerFailed(scaleIn, job, msg)
		return nil
	}

	// wait pods deleted
	deleted, _ := r.isWorkersDeleted(job.Namespace, scaleIn.GetPodNames())
	if deleted {
		job.Status.TargetWorkers = r.workersAfterScaler(job.Status.TargetWorkers, scaleIn)
		job.Status.CurrentWorkers = currentWorkers
		r.updateScalerSuccessd(scaleIn, job)
		return nil
	}

	if remainWorkers {
		msg := "wait for workers process shutdown"
		logger.Info(msg)
		return NewRequeueError(fmt.Errorf(msg))
	}

	return nil
}

6.5 setsSaleInToDelete

通过 ScaleIn CR 中的 spec.toDelete.count 或 spec.toDelete.podNames 字段指定缩容的 worker。

func (r *TrainingJobReconciler) setsSaleInToDelete(job *kaiv1alpha1.TrainingJob, scaleIn *kaiv1alpha1.ScaleIn) error {
	podNames := scaleIn.Status.ToDeletePods
	if len(podNames) != 0 {
		return /*filterPodNames(workers, podNames, false), */ nil
	}
	workers, err := r.GetWorkerPods(job)

	toDelete := scaleIn.Spec.ToDelete

	if toDelete.PodNames != nil {
		workers = filterPodNames(workers, toDelete.PodNames, false)
	} else if toDelete.Count > 0 {
		if toDelete.Count < len(workers) {
			allPodNames := getSortPodNames(job.Name, workers)
			deletePodNames := allPodNames[len(workers)-toDelete.Count:]
			workers = filterPodNames(workers, deletePodNames, false)
		} 
	} 
  
	for _, worker := range workers {
		scaleIn.Status.ToDeletePods = append(scaleIn.Status.ToDeletePods, worker.Name)
	}

	return nil
}


6.6 DeleteWorkers

具体删除worker service 和 pods。

func (r *TrainingJobReconciler) DeleteWorkers(trainingJob *kaiv1alpha1.TrainingJob, workers []string) error {
	if err := r.DeleteWorkerServices(trainingJob, workers); err != nil {
		return fmt.Errorf("delete services failed: %++v", err)
	}

	if err := r.DeleteWorkerPods(trainingJob, workers); err != nil {
		return fmt.Errorf("delete pods failed: %++v", err)
	}
	return nil
}

6.7 DeleteWorkerPods

删除pods。

func (r *TrainingJobReconciler) DeleteWorkerPods(job *kaiv1alpha1.TrainingJob, pods []string) error {
	workerPods, err := r.GetWorkerPods(job)

	if pods != nil {
		workerPods = filterPodNames(workerPods, pods, false)
	}
	for _, pod := range workerPods {
		deleteOptions := &client.DeleteOptions{GracePeriodSeconds: utilpointer.Int64Ptr(0)}
		if err := r.Delete(context.Background(), &pod, deleteOptions); err != nil && !errors.IsNotFound(err) {
			r.recorder.Eventf(job, corev1.EventTypeWarning, trainingJobFailedReason, "Error deleting worker %s: %v", pod.Name, err)
			//return err
		}
		r.recorder.Eventf(job, corev1.EventTypeNormal, trainingJobSucceededReason, "Deleted pod %s", pod.Name)
	}
	return nil
}

具体逻辑如下:

      1 Request("")
 K8S-----------------> Reconcile  <------------------+
      2 ScaleOut CR        +                         |
 K8S----------------->     |                         |
                           |                         |
                           v                         |
    +----------------------+---------------------+   |
    |                 ReconcileJobs              |   |
    |                      +                     |   |
    |                      |                     |   |
    |        +------------------------------+    |   |
    |     1  |             | 2            3 |    |   |
    |        v             v                v    |   | 3
    |  "", JobCreated   JobRunning      Scaling +---------> executeScaling -----+
    +--------+-------------+---------------------+   |          +               |
             |             |                         |          |               |
          1  |             | 2                       |          | 3             | 4
             v             v                         |          v               v
     reconcileResource   reconcileJobRunning         |    executeScaleOut  executeScaleIn
             +             +                         |          +               +
          1  |             | 2                       |          |               |
             |             |                         |          | 3             | 4
             v             v                         |          v               v
+------------+--------+  setTrainingJobScaler        | executeScaleScript executeScaleScript
| doSteps             |    +                         |          +               +
|                     |    | 2                       |          |               |
|                     |    |                         |          | 3             | 4
|    WorkersCreated   |    v                         |          v               v
|                     |  updateScalerState           | hostfileUpdateScript  DeleteWorkers
|                     |    +                         |          +               +
|    WorkersReady     |    |                         |          | 3             | 4
|                     |    | 2                       |          |               |
|                     |    v                         |          v               v
|    LauncherCreated  |  common.Scaling              |   executeOnLauncher  DeleteWorkerPods
|                     |    +                         |          +               +
|                     |    |                         |          |               |
|    JobRunning       |    | 2                       |          | 3             | 4
|                     |    |                         |          v               v
+---------------------+    +-------------------------+     kubectlOnPod      Delete


至此,Horovod系列分析完毕,下一篇开始分析参数服务器,敬请期待。

0xEE 个人信息

★★★★★★关于生活和技术的思考★★★★★★

微信公众账号:罗西的思考

如果您想及时得到个人撰写文章的消息推送,或者想看看个人推荐的技术资料,敬请关注。

在这里插入图片描述

0xFF 参考

ElasticDL 分析

在 Kubernetes 上弹性深度学习训练利器 -- Elastic Training Operator



这篇关于[源码解析] 深度学习分布式训练框架 horovod (20) --- Elastic Training Operator的文章就介绍到这儿,希望我们推荐的文章对大家有所帮助,也希望大家多多支持为之网!


扫一扫关注最新编程教程