Organizational and line-edits include:

- Moved and revised summary para. of 2 versions to intro
- Added internal TOC for user nav
- Simplified Prerequisites
- Revise para. as numbered steps in TensorFlow single and multi-node
- Revised for consistent use of multi-node
- Added URIs for TensorFlow and KubeFlow

Signed-off-by: Michael Vincerra <michael.vincerra@intel.com>
This commit is contained in:
Michael Vincerra
2018-12-06 11:32:08 -08:00
parent 09ee3d9b0f
commit 5ae7faa7e3
+44 -37
View File
@@ -3,22 +3,35 @@
Deep Learning as a Service
##########################
This tutorial explains how to run benchmarking workloads in |CL-ATTR| using
TensorFlow* and Kubeflow with the Intel® Deep Learning Stack.
This tutorial shows you how to run benchmarking workloads in |CL-ATTR| using
TensorFlow\* and Kubeflow with the Intel® Deep Learning Stack.
The Intel® Deep Learning Stack is available in two versions. The first is
`Eigen`_, which includes `TensorFlow`_ optimized for Intel® architecture. The
second is `Intel MKL`_, which includes the TensorFlow framework optimized
using Intel® Math Kernel Library for Deep Neural Networks (Intel® MKL-DNN)
primitives.
.. contents:: :local:
:depth: 1
Release notes
=============
View current `release notes`_ for the Intel® Deep Learning Stack.
.. note::
Performance test numbers in the Intel® Deep Learning Stack were obtained using `runc` as the runtime.
Prerequisites
=============
* |CL| is installed on host system. If not, :ref:`bare-metal-install`
* `containers-basic` bundle is installed
* `cloud-native-basic` bundle is installed
* |CL| installed on host system. If not installed, :ref:`bare-metal-install`
* `containers-basic` bundle
* `cloud-native-basic` bundle
In |CL|, `containers-basic` provides Docker*, which is required for
In |CL|, `containers-basic` provides Docker\*, which is required for
TensorFlow benchmarking. Use the :command:`swupd` utility to check if
`containers-basic` and `cloud-native-basic` are present:
@@ -32,48 +45,41 @@ If you need to install the `containers-basic` or `cloud-native-basic`, enter:
sudo swupd bundle-add containers-basic cloud-native-basic
.. TODO: Resolve issue: Are we installing kubernetes in the bare metal or in the Virtual environment. F-up with DnPlas, bd-dean
To ensure that kubernetes is correctly installed and configured,
:ref:`kubernetes`.
We have validated these steps against the following software package versions
We have validated these steps against the following software package
versions:
* |CL| 26240--lowest version permissible.
* Docker 18.06.1
* Kubernetes 1.11.3
* Go 1.11.12
The Intel® Deep Learning Stack is available in two versions. First, a
version that includes TensorFlow* optimized for Intel Architecture, the
`Eigen`_ version. Second, there is a version that includes the TensorFlow*
framework optimized using Intel® Math Kernel Library for Deep Neural
Networks (Intel® MKL-DNN) primitives, the `Intel MKL`_ version.
TensorFlow* Single and Multi Node Benchmarks
TensorFlow single and multi-node benchmarks
============================================
This section describes running the `TensorFlow benchmarks`_ in single node.
For multi node testing replicate these steps for each node. These steps
provide a template to run other benchmarks, providing they can invoke
For multi-node testing, replicate these steps for each node. These steps
provide a template to run other benchmarks, provided that they can invoke
TensorFlow.
Download and run either the `Eigen`_ or the `Intel MKL-DNN`_ docker image
from hub.docker.com. The next commands will take place in the running
container. Replace <docker_name> with the name of the image.
#. Download and run either the `Eigen`_ or the `Intel MKL-DNN`_ docker image
from `Docker Hub`_.
.. note::
.. note::
You will enter the following commands in the running container.
Replace <docker_name> with the name of the image.
Replace <docker_name> with the name of the image.
#. First, clone the benchmark repository:
#. Clone the benchmark repository:
.. code-block:: bash
docker exec -t <docker_name> bash -c ‘git clone http://github.com/tensorflow/benchmarks -b cnn_tf_v1.11_compatible’
#. Next, execute the benchmark script to run the benchmark
#. Next, execute the benchmark script to run the benchmark.
.. code-block:: bash
@@ -84,11 +90,11 @@ container. Replace <docker_name> with the name of the image.
You can replace the model with one of your choice supported by the
TensorFlow benchmarks.
Kubeflow Multinode benchmarks
Kubeflow multi-node benchmarks
=============================
The benchmark workload will run in a Kubernetes container. We will use
Kubeflow and deploy three nodes for this tutorial to show resource
`Kubeflow`_ and deploy three nodes for this tutorial to show resource
management and get sufficient output data for evaluation.
Kubernetes setup
@@ -103,7 +109,7 @@ Follow the instructions in the :ref:`kubernetes` tutorial to get set up on
Kubernetes networking
*********************
We have used `flannel`_ as the network provider for these tests. If you are
We used `flannel`_ as the network provider for these tests. If you are
comfortable with another network layer, refer to the Kubernetes
`networking documentation`_ for setup.
@@ -111,9 +117,9 @@ comfortable with another network layer, refer to the Kubernetes
Images
******
We need to add `launcher.py` to our docker image to
include the Intel® Deep Learning Stack, and put the benchmarks repo in the
right location. From the docker image, run the following:
We need to add `launcher.py` to our docker image to include the Intel® Deep
Learning Stack and put the benchmarks repo in the correct location. From the
docker image, run the following:
.. code-block:: bash
@@ -125,12 +131,12 @@ right location. From the docker image, run the following:
Your entry point now becomes "/opt/launcher.py".
This will build an image which can be consumed directly by TFJob from
kubeflow. We are working to create these images as part of our release
kubeflow. We are working to create these images as part of our release
cycle.
ksonnet*
********
ksonnet\*
*********
Kubeflow uses ksonnet* to manage deployments, so we need to install that before setting up Kubeflow. On |CL|, follow these steps:
@@ -149,7 +155,7 @@ accessible across the environment.
Kubeflow
********
Once you have Kubernetes running on your nodes, you can setup Kubeflow by following these instructions from their `quick start guide`_.
Once you have Kubernetes running on your nodes, you can setup `Kubeflow`_ by following these instructions from their `quick start guide`_.
.. code-block:: bash
@@ -212,8 +218,9 @@ benchmark results. More information about `Kubernetes logging`_ is available fro
.. The downstream dockerfile will generate another image with benchmarks repo and launcher.py file in the right locations.
.. Dynamic generation of ksonnet template files for a matrix of batch_size, model and replicas.
.. TensorFlow: https://www.tensorflow.org/
.. _Kubeflow: https://www.kubeflow.org/
.. _Docker Hub: https://hub.docker.com/
.. _TensorFlow benchmarks: https://www.tensorflow.org/guide/performance/benchmarks
.. _instructions for creating a cluster: https://kubernetes.io/docs/setup/independent/create-cluster-kubeadm/
.. _flannel: https://github.com/coreos/flannel