README.md 11.4 KB
Newer Older
1
# MindSpore Operator
L
leonwanghui 已提交
2 3 4

#### Experimental notice: This project is still experimental and only serves as a proof of concept for running MindSpore on Kubernetes. The current version of ms-operator is based on an early version of [PyTorch Operator](https://github.com/kubeflow/pytorch-operator) and [TF Operator](https://github.com/kubeflow/tf-operator). Right now MindSpore supports running LeNet with MNIST dataset on a single node, distributed training examples are expected in the near future.

5 6 7 8 9 10 11 12 13 14 15
- [MindSpore Operator](#mindspore-operator)
  - [Introduction of MindSpore and ms-operator](#introduction-of-mindspore-and-ms-operator)
    - [MindSpore Docker Image](#mindspore-docker-image)
    - [Design](#Design)
    - [Overview of MindSpore in Kubeflow Ecosystem](#overview-of-mindspore-in-kubeflow-ecosystem)
  - [Getting Started](#getting-started)
    - [Prerequisites](#prerequisites)
    - [Steps of running the example](#steps-of-running-the-example)
  - [Future Work](#future-work)
  - [Appendix: Example yaml file](#appendix:-example-yaml-file)

L
leonwanghui 已提交
16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
## Introduction of MindSpore and ms-operator

MindSpore is a new open source deep learning training/inference framework that
could be used for mobile, edge and cloud scenarios. MindSpore is designed to
provide development experience with friendly design and efficient execution for
the data scientists and algorithmic engineers, native support for Ascend AI
processor, and software hardware co-optimization.

This project contains the specification and implementation of MSJob custom
resource definition. We will demonstrate running a walkthrough of creating
ms-operator, as well as MNIST training job on Kubernetes with MindSpore
0.1.0-alpha image (x86 CPU build version) on a single node. More completed
features will be developed in the coming days.

This project defines the following:
- The ms-operator
- A way to deploy the operator
- MindSpore LeNet MNIST training example
- Future goal: distributed MindSpore training example

### MindSpore Docker Image

38 39 40 41
MindSpore docker image is hosted on [Docker Hub](https://hub.docker.com/r/mindspore), currently both `CPU` and `GPU` are supported as follows:
- CPU: `mindspore/mindspore-cpu:0.1.0-alpha`
- GPU (Cuda 10.1): `mindspore/mindspore-cuda10.1:0.1.0-alpha`
- GPU (Cuda 9.2): `mindspore/mindspore-cuda9.2:0.1.0-alpha`
L
leonwanghui 已提交
42

43
Take CPU for example, you can directly pull the image using the below command:
L
leonwanghui 已提交
44
```
45
docker pull mindspore/mindspore-cpu:0.1.0-alpha
L
leonwanghui 已提交
46 47 48 49 50 51 52 53 54 55 56 57
```

### Design

The yaml file we used to create our MNIST training job is defined as follows:
```yaml
apiVersion: v1
kind: Pod
metadata:
  name: msjob-mnist
spec:
  containers:
58
  - image: mindspore/mindspore-cpu:0.1.0-alpha
L
leonwanghui 已提交
59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77
    imagePullPolicy: IfNotPresent
    name: msjob-mnist
    command: ["/bin/bash", "-c", "python /tmp/test/MNIST/lenet.py"]
    volumeMounts:
      - name: training-result
        mountPath: /tmp/result
      - name: ms-mnist
        mountPath: /tmp/test
  restartPolicy: OnFailure
  volumes:
    - name: training-result
      emptyDir: {}
    - name: ms-mnist
      hostPath:
        path: /root/gopath/src/gitee.com/mindspore/ms-operator/examples/
```

### Overview of MindSpore in Kubeflow Ecosystem

78
<img src="./docs/pics/ms-operator-in-kubeflow.png" alt="ms-operator in Kubeflow" width=600/>
L
leonwanghui 已提交
79 80 81 82

The high-level view of how MindSpore fits in the ecosystem of Kubeflow and its
components.

83
## Getting Started
L
leonwanghui 已提交
84

85
### Prerequisites
L
leonwanghui 已提交
86

87 88 89 90
- [Helm and Tiller](https://github.com/helm/helm/releases/tag/v2.9.0): `v2.9.0`
- [go](https://github.com/golang/go/releases/tag/go1.12.1): `go1.12.1`
- [docker](https://github.com/docker/docker-ce/releases/tag/v18.06.1-ce): `v18.06.1-ce`
- [Kubernetes](https://github.com/kubernetes/kubernetes/releases/tag/v1.14.0): `v1.14.0`
L
leonwanghui 已提交
91

92
### Steps of running the example
L
leonwanghui 已提交
93

94
First, pull the ms-operator image from [Docker Hub](https://hub.docker.com/r/mindspore):
L
leonwanghui 已提交
95
```
96
docker pull mindspore/ms-operator:latest
L
leonwanghui 已提交
97 98
```

99 100 101 102
Or you build the ms-operator image on local machine:
```
docker build . -t mindspore/ms-operator
```
L
leonwanghui 已提交
103

104
After the installation, check the image status using `docker images` command:
L
leonwanghui 已提交
105
```
106 107
REPOSITORY                        TAG                   IMAGE ID            CREATED             SIZE
mindspore/ms-operator             latest                729960ae415e        28 hours ago        175MB
L
leonwanghui 已提交
108 109 110 111
```

The MindSpore image we download from docker hub is `0.1.0-alpha` version:
```
112 113
REPOSITORY                        TAG                   IMAGE ID            CREATED             SIZE
mindspore/mindspore-cpu           0.1.0-alpha           9a124f33ed27        2 hours ago         1.19GB
L
leonwanghui 已提交
114 115 116 117 118 119 120 121
```

MindSpore supports heterogeneous computing including multiple hardware and
backends (`CPU`, `GPU`, `Ascend`), the device_target of MindSpore is
`Ascend` by default but we will use the CPU version here.

Install the msjob crd, ms-operator deployment and pod:
```
122
RBAC=true # set false if you do not have an RBAC cluster
L
leonwanghui 已提交
123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176
helm install ms-operator-chart/ -n ms-operator --set rbac.install=${RBAC} --wait --replace
```

Using `helm status ms-operator` command to check generated resources:
```
LAST DEPLOYED: Tue Mar 24 11:36:51 2020
NAMESPACE: default
STATUS: DEPLOYED

RESOURCES:
==> v1beta1/CustomResourceDefinition
NAME                 AGE
msjobs.kubeflow.org  1d

==> v1beta1/Deployment
NAME         DESIRED  CURRENT  UP-TO-DATE  AVAILABLE  AGE
ms-operator  1        1        1           1          1d

==> v1/Pod(related)
NAME                          READY  STATUS   RESTARTS  AGE
ms-operator-7b5b457d69-dpd2b  1/1    Running  0         1d
```

We will do a MNIST training to check the eligibility of MindSpore running on
Kubernetes:
```
cd examples/ && kubectl apply -f ms-mnist.yaml
```

The job is simply importing MindSpore packges, the dataset is already included in the `MNIST_Data` folder, executing only one epoch and printing result which should only consume little time. After the job completed, you should be able to check the job status and see the result logs. You can check the source training code in `examples/` folder.
```
kubectl get pod msjob-mnist && kubectl logs msjob-mnist
```
```
NAME          READY   STATUS      RESTARTS   AGE
msjob-mnist   0/1     Completed   0          3h53m
============== Starting Training ==============
epoch: 1 step: 1, loss is 2.3005836
epoch: 1 step: 2, loss is 2.2978227
epoch: 1 step: 3, loss is 2.3004227
epoch: 1 step: 4, loss is 2.3054247
epoch: 1 step: 5, loss is 2.3068798
epoch: 1 step: 6, loss is 2.298408
epoch: 1 step: 7, loss is 2.3055573
epoch: 1 step: 8, loss is 2.2998955
epoch: 1 step: 9, loss is 2.3028255
epoch: 1 step: 10, loss is 2.2972553
```

Since MindSpore is in the early stage of open source, the whole community is
still working on implementing distributed training of LeNet with MNIST dataset
on Kubernetes, together with the distributed training on different backends
(GPU || `Ascend`) are also expected in the near future.

177
## Future Work
L
leonwanghui 已提交
178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214

[Kubeflow](https://github.com/kubeflow/kubeflow) just announced its first major
1.0 release recently with the graduation of a core set of stable applications
including:
- [Kubeflow's UI](https://www.kubeflow.org/docs/components/central-dash/overview/)
- [Jupyter notebook controller](https://github.com/kubeflow/kubeflow/tree/master/components/notebook-controller) and [web app](https://www.kubeflow.org/docs/notebooks/why-use-jupyter-notebook/)
- [Tensorflow Operator](https://www.kubeflow.org/docs/components/training/tftraining/)(TFJob), and [PyTorch Operator](https://www.kubeflow.org/docs/components/training/pytorch/) for distributed training
- [kfctl](https://www.kubeflow.org/docs/other-guides/kustomize/) for deployment and upgrade
- etc.

The MindSpore community is driving to collaborate with the Kubeflow community
as well as making the ms-operator more complex, well-orgnized and its 
dependencies up-to-date. All these components make it easy for machine learning
engineers and data scientists to leverage cloud assets (public or on-premise)
for machine learning workloads.

MindSpore is also looking forward to enable users to use Jupyter to develop
models. Users in the future can use Kubeflow tools like fairing (Kubeflow’s
python SDK) to build containers and create Kubernetes resources to train their
MindSpore models.

Once training completed, users can use [KFServing](https://github.com/kubeflow/kfserving)
to create and deploy a server for inference thus completing the life cycle of
machine learning.

Distributed training is another field MindSpore will be focusing on. There are
two major distributed training strategies nowadays: one based on parameter
servers and the other based on collective communication primitives such as
allreduce. [MPI Operator](https://github.com/kubeflow/mpi-operator) is one of
the core components of Kubeflow which makes it easy to run synchronized,
allreduce-style distributed training on Kubernetes. MPI Operator provides a crd
for defining a training job on a single CPU/GPU, multiple CPU/GPUs, and multiple
nodes. It also implements a custom controller to manage the CRD, create
dependent resources, and reconcile the desired states. If MindSpore can leverage
MPI Operator together with the high performance `Ascend` processor, it is
possible that MindSpore will bring distributed training to an even higher level.

215
## Appendix: Example yaml file
L
leonwanghui 已提交
216

217
The yaml file to create distributed training MSJob expected to be like this:
L
leonwanghui 已提交
218 219 220 221 222 223 224 225 226 227 228 229 230 231 232
```yaml
# WIP example for distributed training
apiVersion: "kubeflow.org/v1"
kind: "MSJob"
metadata:
  name: "msjob-mnist"
spec:
  backend: "tcp"
  masterPort: "23456"
  replicaSpecs:
    - replicas: 1
      replicaType: MASTER
      template:
        spec:
          containers:
233
          - image: mindspore/mindspore-cpu:0.1.0-alpha
L
leonwanghui 已提交
234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255
            imagePullPolicy: IfNotPresent
            name: msjob-mnist
            command: ["/bin/bash", "-c", "python /tmp/test/MNIST/lenet.py"]
            volumeMounts:
              - name: training-result
                mountPath: /tmp/result
              - name: ms-mnist-local-file
                mountPath: /tmp/test
          restartPolicy: OnFailure
          volumes:
            - name: training-result
              emptyDir: {}
            - name: entrypoint
              configMap:
                name: dist-train
                defaultMode: 0755
          restartPolicy: OnFailure
    - replicas: 3
      replicaType: WORKER
      template:
        spec:
          containers:
256
          - image: mindspore/mindspore-cpu:0.1.0-alpha
L
leonwanghui 已提交
257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285
            imagePullPolicy: IfNotPresent
            name: msjob-mnist
            command: ["/bin/bash", "-c", "python /tmp/test/MNIST/lenet.py"]
            volumeMounts:
              - name: training-result
                mountPath: /tmp/result
              - name: ms-mnist-local-file
                hostPath:
                    path: /root/gopath/src/gitee.com/mindspore/ms-operator/examples
          restartPolicy: OnFailure
          volumes:
            - name: training-result
              emptyDir: {}
            - name: entrypoint
              configMap:
                name: dist-train
                defaultMode: 0755
          restartPolicy: OnFailure
```

The MSJob currently is designed based on the TF Job and PyTorch Job,
and is subject to change in future versions.

We define `backend` protocol which the MS workers will use to communicate when
initializing the worker group. MindSpore supports heterogeneous computing
including multiple hardware and backends (`CPU`, `GPU`, `Ascend`),
the device_target of MindSpore is `Ascend` by default.

We define `masterPort` that groups will use to communicate with master service.