ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

Update k8s tutorial and script (#7356)

Browse source 
* debug

* update train

* update remote dir

* remove print

* update train

* update
This commit is contained in:
Kai Huang 2023-02-06 11:24:07 +08:00 committed by GitHub
parent 4b8b6b7d0e
commit 8911c69302
2 changed files with 16 additions and 16 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

18
docs/readthedocs/source/doc/Orca/Tutorial/k8s.md
View file

@ -234,7 +234,7 @@ cp /path/to/fashion-mnist/data/fashion/* /bigdl/nfsdata/dataset/FashionMNIST/raw
gzip -d /bigdl/nfsdata/dataset/FashionMNIST/raw/* gzip -d /bigdl/nfsdata/dataset/FashionMNIST/raw/*
``` ```
In the given example, you can specify the argument `--remote_dir` to be the directory on NFS for the Fashion-MNIST dataset. The directory should contain `FashionMNIST/raw/train-images-idx3-ubyte` and `FashionMNIST/raw/t10k-images-idx3`. In the given example, you can specify the argument `--data_dir` to be the directory on NFS for the Fashion-MNIST dataset. The directory should contain `FashionMNIST/raw/train-images-idx3-ubyte` and `FashionMNIST/raw/t10k-images-idx3`.
--- ---
@ -304,7 +304,7 @@ See [here](#init-orca-context) for the runtime configurations.
#### 6.1.1 K8s-Client #### 6.1.1 K8s-Client
Run the example with the following command by setting the cluster_mode to "k8s-client": Run the example with the following command by setting the cluster_mode to "k8s-client":
```bash ```bash
python train.py --cluster_mode k8s-client --remote_dir file:///bigdl/nfsdata/dataset python train.py --cluster_mode k8s-client --data_dir /bigdl/nfsdata/dataset
``` ```
@ -333,7 +333,7 @@ conda pack -o environment.tar.gz
Run the example with the following command by setting the cluster_mode to "k8s-cluster": Run the example with the following command by setting the cluster_mode to "k8s-cluster":
```bash ```bash
python /bigdl/nfsdata/train.py --cluster_mode k8s-cluster --remote_dir /bigdl/nfsdata/dataset python /bigdl/nfsdata/train.py --cluster_mode k8s-cluster --data_dir /bigdl/nfsdata/dataset
``` ```
@ -395,7 +395,7 @@ ${SPARK_HOME}/bin/spark-submit \
--conf spark.executor.extraClassPath=${BIGDL_HOME}/jars/* \ --conf spark.executor.extraClassPath=${BIGDL_HOME}/jars/* \
--conf spark.kubernetes.executor.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.options.claimName=${RUNTIME_PERSISTENT_VOLUME_CLAIM} \ --conf spark.kubernetes.executor.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.options.claimName=${RUNTIME_PERSISTENT_VOLUME_CLAIM} \
--conf spark.kubernetes.executor.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.mount.path=/bigdl/nfsdata \ --conf spark.kubernetes.executor.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.mount.path=/bigdl/nfsdata \
train.py --cluster_mode spark-submit --remote_dir /bigdl/nfsdata/dataset train.py --cluster_mode spark-submit --data_dir /bigdl/nfsdata/dataset
``` ```
In the `spark-submit` script: In the `spark-submit` script:
@ -447,7 +447,7 @@ ${SPARK_HOME}/bin/spark-submit \
--conf spark.kubernetes.driver.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.mount.path=/bigdl/nfsdata \ --conf spark.kubernetes.driver.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.mount.path=/bigdl/nfsdata \
--conf spark.kubernetes.executor.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.options.claimName=${RUNTIME_PERSISTENT_VOLUME_CLAIM} \ --conf spark.kubernetes.executor.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.options.claimName=${RUNTIME_PERSISTENT_VOLUME_CLAIM} \
--conf spark.kubernetes.executor.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.mount.path=/bigdl/nfsdata \ --conf spark.kubernetes.executor.volumes.persistentVolumeClaim.${RUNTIME_PERSISTENT_VOLUME_CLAIM}.mount.path=/bigdl/nfsdata \
file:///bigdl/nfsdata/train.py --cluster_mode spark-submit --remote_dir /bigdl/nfsdata/dataset file:///bigdl/nfsdata/train.py --cluster_mode spark-submit --data_dir /bigdl/nfsdata/dataset
``` ```
In the `spark-submit` script: In the `spark-submit` script:
@ -544,7 +544,7 @@ spec:
--conf spark.executor.extraClassPath=local://${BIGDL_HOME}/jars/* \ --conf spark.executor.extraClassPath=local://${BIGDL_HOME}/jars/* \
local:///bigdl/nfsdata/train.py local:///bigdl/nfsdata/train.py
--cluster_mode spark-submit --cluster_mode spark-submit
--remote_dir file:///bigdl/nfsdata/dataset --data_dir file:///bigdl/nfsdata/dataset
"] "]
securityContext: securityContext:
privileged: true privileged: true
@ -688,7 +688,7 @@ spec:
--conf spark.executor.extraClassPath=local://${BIGDL_HOME}/jars/* \ --conf spark.executor.extraClassPath=local://${BIGDL_HOME}/jars/* \
local:///bigdl/nfsdata/train.py local:///bigdl/nfsdata/train.py
--cluster_mode spark-submit --cluster_mode spark-submit
--remote_dir file:///bigdl/nfsdata/dataset --data_dir file:///bigdl/nfsdata/dataset
"] "]
securityContext: securityContext:
privileged: true privileged: true
@ -861,7 +861,7 @@ spec:
--conf spark.executor.extraClassPath=local://${BIGDL_HOME}/jars/* \ --conf spark.executor.extraClassPath=local://${BIGDL_HOME}/jars/* \
local:///bigdl/nfsdata/train.py local:///bigdl/nfsdata/train.py
--cluster_mode spark-submit --cluster_mode spark-submit
--remote_dir file:///bigdl/nfsdata/dataset --data_dir file:///bigdl/nfsdata/dataset
"] "]
securityContext: securityContext:
privileged: true privileged: true
@ -999,7 +999,7 @@ spec:
--conf spark.executor.extraClassPath=local://${BIGDL_HOME}/jars/* \ --conf spark.executor.extraClassPath=local://${BIGDL_HOME}/jars/* \
local:///bigdl/nfsdata/train.py local:///bigdl/nfsdata/train.py
--cluster_mode spark-submit --cluster_mode spark-submit
--remote_dir file:///bigdl/nfsdata/dataset --data_dir file:///bigdl/nfsdata/dataset
"] "]
securityContext: securityContext:
privileged: true privileged: true

14
docs/readthedocs/source/doc/Orca/Tutorial/yarn.md
View file

@ -127,7 +127,7 @@ Then upload it to a distributed storage. Sample command to upload data to HDFS i
```bash ```bash
hdfs dfs -put /path/to/local/data/FashionMNIST hdfs://path/to/remote/data hdfs dfs -put /path/to/local/data/FashionMNIST hdfs://path/to/remote/data
``` ```
In the given example, you can specify the argument `--remote_dir` to be the directory on a distributed storage for the Fashion-MNIST dataset. The directory should contain `FashionMNIST/raw/train-images-idx3-ubyte` and `FashionMNIST/raw/t10k-images-idx3`. In the given example, you can specify the argument `--data_dir` to be the directory on a distributed storage for the Fashion-MNIST dataset. The directory should contain `FashionMNIST/raw/train-images-idx3-ubyte` and `FashionMNIST/raw/t10k-images-idx3`.
--- ---
## 4. Prepare Custom Modules ## 4. Prepare Custom Modules
@ -192,14 +192,14 @@ See [here](#init-orca-context) for the runtime configurations.
#### 5.1.1 Yarn Client #### 5.1.1 Yarn Client
Run the example with the following command by setting the cluster_mode to "yarn-client": Run the example with the following command by setting the cluster_mode to "yarn-client":
```bash ```bash
python train.py --cluster_mode yarn-client --remote_dir hdfs://path/to/remote/data python train.py --cluster_mode yarn-client --data_dir hdfs://path/to/remote/data
``` ```
#### 5.1.2 Yarn Cluster #### 5.1.2 Yarn Cluster
Run the example with the following command by setting the cluster_mode to "yarn-cluster": Run the example with the following command by setting the cluster_mode to "yarn-cluster":
```bash ```bash
python train.py --cluster_mode yarn-cluster --remote_dir hdfs://path/to/remote/data python train.py --cluster_mode yarn-cluster --data_dir hdfs://path/to/remote/data
``` ```
@ -254,7 +254,7 @@ bigdl-submit \
--archives /path/to/environment.tar.gz#environment \ --archives /path/to/environment.tar.gz#environment \
--conf spark.pyspark.driver.python=/path/to/python \ --conf spark.pyspark.driver.python=/path/to/python \
--conf spark.pyspark.python=environment/bin/python \ --conf spark.pyspark.python=environment/bin/python \
train.py --cluster_mode bigdl-submit --remote_dir hdfs://path/to/remote/data train.py --cluster_mode bigdl-submit --data_dir hdfs://path/to/remote/data
``` ```
In the `bigdl-submit` script: In the `bigdl-submit` script:
* `--master`: the spark master, set it to "yarn". * `--master`: the spark master, set it to "yarn".
@ -277,7 +277,7 @@ bigdl-submit \
--archives /path/to/environment.tar.gz#environment \ --archives /path/to/environment.tar.gz#environment \
--conf spark.yarn.appMasterEnv.PYSPARK_PYTHON=environment/bin/python \ --conf spark.yarn.appMasterEnv.PYSPARK_PYTHON=environment/bin/python \
--conf spark.executorEnv.PYSPARK_PYTHON=environment/bin/python \ --conf spark.executorEnv.PYSPARK_PYTHON=environment/bin/python \
train.py --cluster_mode bigdl-submit --remote_dir hdfs://path/to/remote/data train.py --cluster_mode bigdl-submit --data_dir hdfs://path/to/remote/data
``` ```
In the `bigdl-submit` script: In the `bigdl-submit` script:
* `--master`: the spark master, set it to "yarn". * `--master`: the spark master, set it to "yarn".
@ -344,7 +344,7 @@ ${SPARK_HOME}/bin/spark-submit \
--conf spark.pyspark.python=environment/bin/python \ --conf spark.pyspark.python=environment/bin/python \
--py-files ${BIGDL_HOME}/python/bigdl-spark_${SPARK_VERSION}-${BIGDL_VERSION}-python-api.zip,model.py \ --py-files ${BIGDL_HOME}/python/bigdl-spark_${SPARK_VERSION}-${BIGDL_VERSION}-python-api.zip,model.py \
--jars ${BIGDL_HOME}/jars/bigdl-assembly-spark_${SPARK_VERSION}-${BIGDL_VERSION}-jar-with-dependencies.jar \ --jars ${BIGDL_HOME}/jars/bigdl-assembly-spark_${SPARK_VERSION}-${BIGDL_VERSION}-jar-with-dependencies.jar \
train.py --cluster_mode spark-submit --remote_dir hdfs://path/to/remote/data train.py --cluster_mode spark-submit --data_dir hdfs://path/to/remote/data
``` ```
In the `spark-submit` script: In the `spark-submit` script:
* `--master`: the spark master, set it to "yarn". * `--master`: the spark master, set it to "yarn".
@ -368,7 +368,7 @@ ${SPARK_HOME}/bin/spark-submit \
--conf spark.executorEnv.PYSPARK_PYTHON=environment/bin/python \ --conf spark.executorEnv.PYSPARK_PYTHON=environment/bin/python \
--py-files ${BIGDL_HOME}/python/bigdl-spark_${SPARK_VERSION}-${BIGDL_VERSION}-python-api.zip,model.py \ --py-files ${BIGDL_HOME}/python/bigdl-spark_${SPARK_VERSION}-${BIGDL_VERSION}-python-api.zip,model.py \
--jars ${BIGDL_HOME}/jars/bigdl-assembly-spark_${SPARK_VERSION}-${BIGDL_VERSION}-jar-with-dependencies.jar \ --jars ${BIGDL_HOME}/jars/bigdl-assembly-spark_${SPARK_VERSION}-${BIGDL_VERSION}-jar-with-dependencies.jar \
train.py --cluster_mode spark-submit --remote_dir hdfs://path/to/remote/data train.py --cluster_mode spark-submit --data_dir hdfs://path/to/remote/data
``` ```
In the `spark-submit` script: In the `spark-submit` script:
* `--master`: the spark master, set it to "yarn". * `--master`: the spark master, set it to "yarn".