Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add MNIST test to run multi-node distributed training using KFTO #295

Merged
merged 4 commits into from
Jan 3, 2025
Merged

Add MNIST test to run multi-node distributed training using KFTO #295

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
4 commits
Select commit Hold shift + click to select a range
9ae3ffd
Add MNIST test to run multi-node distributed training using CPUs or N…
abhijeet-dhumal Dec 16, 2024
c5686c8
updated MNIST test to use antipod-affinity for scheduling worker node…
abhijeet-dhumal Dec 19, 2024
4b92b1d
Add inter-pod anti-affinity rule in master node spec to schedule node…
abhijeet-dhumal Dec 19, 2024
7dbfdfc
Added resources list for CPU based usecase, updated requirements.txt …
abhijeet-dhumal Dec 23, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
306 changes: 306 additions & 0 deletions tests/kfto/kfto_mnist_training_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,306 @@
/*
Copyright 2023.

Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
*/

package kfto

import (
"bytes"
"fmt"
"testing"

kftov1 "github.com/kubeflow/training-operator/pkg/apis/kubeflow.org/v1"
. "github.com/onsi/gomega"
. "github.com/project-codeflare/codeflare-common/support"

corev1 "k8s.io/api/core/v1"
"k8s.io/apimachinery/pkg/api/resource"
metav1 "k8s.io/apimachinery/pkg/apis/meta/v1"
)

func TestPyTorchJobMnistCpu(t *testing.T) {
runKFTOPyTorchMnistJob(t, 0, 2, "", GetCudaTrainingImage(), "resources/requirements.txt")
}
func TestPyTorchJobMnistWithCuda(t *testing.T) {
runKFTOPyTorchMnistJob(t, 1, 1, "nvidia.com/gpu", GetCudaTrainingImage(), "resources/requirements.txt")
}

func TestPyTorchJobMnistWithROCm(t *testing.T) {
runKFTOPyTorchMnistJob(t, 1, 1, "amd.com/gpu", GetROCmTrainingImage(), "resources/requirements-rocm.txt")
}

func runKFTOPyTorchMnistJob(t *testing.T, numGpus int, workerReplicas int, gpuLabel string, image string, requirementsFile string) {
abhijeet-dhumal marked this conversation as resolved.
Show resolved Hide resolved
abhijeet-dhumal marked this conversation as resolved.
Show resolved Hide resolved
test := With(t)

// Create a namespace
namespace := test.NewTestNamespace()

mnist := ReadFile(test, "resources/mnist.py")
requirementsFileName := ReadFile(test, requirementsFile)

if numGpus > 0 {
mnist = bytes.Replace(mnist, []byte("accelerator=\"has to be specified\""), []byte("accelerator=\"gpu\""), 1)
} else {
mnist = bytes.Replace(mnist, []byte("accelerator=\"has to be specified\""), []byte("accelerator=\"cpu\""), 1)
}
config := CreateConfigMap(test, namespace.Name, map[string][]byte{
// MNIST Ray Notebook
"mnist.py": mnist,
"requirements.txt": requirementsFileName,
})

outputPvc := CreatePersistentVolumeClaim(test, namespace.Name, "50Gi", corev1.ReadWriteOnce)
defer test.Client().Core().CoreV1().PersistentVolumeClaims(namespace.Name).Delete(test.Ctx(), outputPvc.Name, metav1.DeleteOptions{})

// Create training PyTorch job
tuningJob := createKFTOPyTorchMnistJob(test, namespace.Name, *config, gpuLabel, numGpus, workerReplicas, outputPvc.Name, image)
defer test.Client().Kubeflow().KubeflowV1().PyTorchJobs(namespace.Name).Delete(test.Ctx(), tuningJob.Name, *metav1.NewDeleteOptions(0))

// Make sure the PyTorch job is running
test.Eventually(PyTorchJob(test, namespace.Name, tuningJob.Name), TestTimeoutDouble).
Should(WithTransform(PyTorchJobConditionRunning, Equal(corev1.ConditionTrue)))

// Make sure the PyTorch job succeeded
test.Eventually(PyTorchJob(test, namespace.Name, tuningJob.Name), TestTimeoutDouble).Should(WithTransform(PyTorchJobConditionSucceeded, Equal(corev1.ConditionTrue)))
test.T().Logf("PytorchJob %s/%s ran successfully", tuningJob.Namespace, tuningJob.Name)

}

func createKFTOPyTorchMnistJob(test Test, namespace string, config corev1.ConfigMap, gpuLabel string, numGpus int, workerReplicas int, outputPvcName string, baseImage string) *kftov1.PyTorchJob {
var useGPU = false
var backend string

if numGpus > 0 {
useGPU = true
backend = "nccl"
} else {
backend = "gloo"
}

tuningJob := &kftov1.PyTorchJob{
TypeMeta: metav1.TypeMeta{
APIVersion: corev1.SchemeGroupVersion.String(),
Kind: "PyTorchJob",
},
ObjectMeta: metav1.ObjectMeta{
GenerateName: "kfto-mnist-",
},
Spec: kftov1.PyTorchJobSpec{
PyTorchReplicaSpecs: map[kftov1.ReplicaType]*kftov1.ReplicaSpec{
"Master": {
Replicas: Ptr(int32(1)),
RestartPolicy: kftov1.RestartPolicyOnFailure,
Template: corev1.PodTemplateSpec{
ObjectMeta: metav1.ObjectMeta{
Labels: map[string]string{
"app": "kfto-mnist",
},
},
Spec: corev1.PodSpec{
Affinity: &corev1.Affinity{
PodAntiAffinity: &corev1.PodAntiAffinity{
RequiredDuringSchedulingIgnoredDuringExecution: []corev1.PodAffinityTerm{
{
LabelSelector: &metav1.LabelSelector{
MatchLabels: map[string]string{
"app": "kfto-mnist",
},
},
TopologyKey: "kubernetes.io/hostname",
},
},
},
},
Containers: []corev1.Container{
{
Name: "pytorch",
Image: baseImage,
ImagePullPolicy: corev1.PullIfNotPresent,
Command: []string{
"/bin/bash", "-c",
fmt.Sprintf(`mkdir -p /tmp/lib && export PYTHONPATH=$PYTHONPATH:/tmp/lib && \
pip install --no-cache-dir -r /mnt/files/requirements.txt --target=/tmp/lib && \
python /mnt/files/mnist.py --epochs 3 --save-model --output-path /mnt/output --backend %s`, backend),
},
VolumeMounts: []corev1.VolumeMount{
{
Name: config.Name,
MountPath: "/mnt/files",
},
{
Name: "tmp-volume",
MountPath: "/tmp",
},
{
Name: "output-volume",
MountPath: "/mnt/output",
},
},
Resources: corev1.ResourceRequirements{
Limits: corev1.ResourceList{
corev1.ResourceCPU: resource.MustParse("1"),
corev1.ResourceMemory: resource.MustParse("6Gi"),
},
},
},
},
Volumes: []corev1.Volume{
{
Name: config.Name,
VolumeSource: corev1.VolumeSource{
ConfigMap: &corev1.ConfigMapVolumeSource{
LocalObjectReference: corev1.LocalObjectReference{
Name: config.Name,
},
},
},
},
{
Name: "tmp-volume",
VolumeSource: corev1.VolumeSource{
EmptyDir: &corev1.EmptyDirVolumeSource{},
},
},
{
Name: "output-volume",
VolumeSource: corev1.VolumeSource{
PersistentVolumeClaim: &corev1.PersistentVolumeClaimVolumeSource{
ClaimName: outputPvcName,
},
},
},
},
RestartPolicy: corev1.RestartPolicyOnFailure,
},
},
},
"Worker": {
Replicas: Ptr(int32(workerReplicas)),
RestartPolicy: kftov1.RestartPolicyOnFailure,
Template: corev1.PodTemplateSpec{
abhijeet-dhumal marked this conversation as resolved.
Show resolved Hide resolved
ObjectMeta: metav1.ObjectMeta{
Labels: map[string]string{
"app": "kfto-mnist",
},
},
Spec: corev1.PodSpec{
Affinity: &corev1.Affinity{
PodAntiAffinity: &corev1.PodAntiAffinity{
RequiredDuringSchedulingIgnoredDuringExecution: []corev1.PodAffinityTerm{
{
LabelSelector: &metav1.LabelSelector{
MatchLabels: map[string]string{
"app": "kfto-mnist",
},
},
TopologyKey: "kubernetes.io/hostname",
},
},
},
},
Containers: []corev1.Container{
{
Name: "pytorch",
Image: baseImage,
ImagePullPolicy: corev1.PullIfNotPresent,
Command: []string{
"/bin/bash", "-c",
fmt.Sprintf(`mkdir -p /tmp/lib && export PYTHONPATH=$PYTHONPATH:/tmp/lib && \
pip install --no-cache-dir -r /mnt/files/requirements.txt --target=/tmp/lib && \
python /mnt/files/mnist.py --epochs 3 --save-model --backend %s`, backend),
},
VolumeMounts: []corev1.VolumeMount{
{
Name: config.Name,
MountPath: "/mnt/files",
},
{
Name: "tmp-volume",
MountPath: "/tmp",
},
},
Resources: corev1.ResourceRequirements{
Limits: corev1.ResourceList{
corev1.ResourceCPU: resource.MustParse("1"),
corev1.ResourceMemory: resource.MustParse("6Gi"),
},
},
},
},
Volumes: []corev1.Volume{
{
Name: config.Name,
VolumeSource: corev1.VolumeSource{
ConfigMap: &corev1.ConfigMapVolumeSource{
LocalObjectReference: corev1.LocalObjectReference{
Name: config.Name,
},
},
},
},
{
Name: "tmp-volume",
VolumeSource: corev1.VolumeSource{
EmptyDir: &corev1.EmptyDirVolumeSource{},
},
},
},
RestartPolicy: corev1.RestartPolicyOnFailure,
},
},
},
},
},
}

if useGPU {
abhijeet-dhumal marked this conversation as resolved.
Show resolved Hide resolved
// Update resource lists for GPU (NVIDIA/ROCm) usecase
tuningJob.Spec.PyTorchReplicaSpecs["Master"].Template.Spec.Containers[0].Resources.Limits[corev1.ResourceName(gpuLabel)] = resource.MustParse(fmt.Sprint(numGpus))
tuningJob.Spec.PyTorchReplicaSpecs["Worker"].Template.Spec.Containers[0].Resources.Limits[corev1.ResourceName(gpuLabel)] = resource.MustParse(fmt.Sprint(numGpus))

tuningJob.Spec.PyTorchReplicaSpecs["Master"].Template.Spec.Containers[0].Env = []corev1.EnvVar{
{
Name: "NCCL_DEBUG",
Value: "INFO",
},
}
tuningJob.Spec.PyTorchReplicaSpecs["Worker"].Template.Spec.Containers[0].Env = []corev1.EnvVar{
{
Name: "NCCL_DEBUG",
Value: "INFO",
},
}

// Update tolerations
tuningJob.Spec.PyTorchReplicaSpecs["Master"].Template.Spec.Tolerations = []corev1.Toleration{
{
Key: gpuLabel,
Operator: corev1.TolerationOpExists,
},
}
tuningJob.Spec.PyTorchReplicaSpecs["Worker"].Template.Spec.Tolerations = []corev1.Toleration{
{
Key: gpuLabel,
Operator: corev1.TolerationOpExists,
},
}
}

tuningJob, err := test.Client().Kubeflow().KubeflowV1().PyTorchJobs(namespace).Create(test.Ctx(), tuningJob, metav1.CreateOptions{})
test.Expect(err).NotTo(HaveOccurred())
test.T().Logf("Created PytorchJob %s/%s successfully", tuningJob.Namespace, tuningJob.Name)

return tuningJob
}
Loading
Loading