Download as PDF, PPTX
![TensorFlow Cluster
Node #0
tf.train.ClusterSpec({
"worker": [
"worker0.example.com:2222",
"worker1.example.com:2222"
],
"ps": [
"ps0.example.com:2222",
"ps1.example.com:2222"
]})
Worker
Node #1
Worker
Node #1
PS
Node #0
PS](https://image.slidesharecdn.com/clusteringtensorflowconkubernetesyraspberrypi-180713063620/75/Clustering-tensor-flow-con-kubernetes-y-raspberry-pi-22-2048.jpg)
![TensorFlow Cluster
Node #0
tf.train.ClusterSpec({
"worker": [
"worker0.example.com:2222",
"worker1.example.com:2222"
],
"ps": [
"ps0.example.com:2222",
"ps1.example.com:2222"
]})
Worker
Node #1
Worker
Node #1
PS
Node #0
PS](https://crownmelresort.com/image.slidesharecdn.com/clusteringtensorflowconkubernetesyraspberrypi-180713063620/75/Clustering-tensor-flow-con-kubernetes-y-raspberry-pi-22-2048.jpg)
Uploaded byAndrés Leonardo Martinez Ortiz
Clustering tensor flow con kubernetes y raspberry pi
This document discusses using Kubernetes to cluster Raspberry Pi devices running TensorFlow. It begins by introducing Kubernetes, TensorFlow, and the Raspberry Pi. It then covers setting up a Kubernetes cluster across multiple Raspberry Pis, including installing Docker, configuring the master and nodes, and deploying networking. Next, it discusses deploying TensorFlow jobs in a distributed manner across the Kubernetes cluster using strategies like in-graph replication. It also proposes using Docker images and Ansible scripts to simplify and automate the cluster setup. Finally, it outlines how the cluster could be used for applications involving hyperparameter tuning, scaling ML APIs, and ensemble/data parallelism with TensorFlow.
More Related Content
![[POSS 2019] OVirt and Ceph: Perfect Combination.?](https://cdn.slidesharecdn.com/ss_thumbnails/pre-poss2019-worteks-ovirt-ceph-191216145947-thumbnail.jpg?width=640&height=640&fit=bounds)
Similar to Clustering tensor flow con kubernetes y raspberry pi
More from Andrés Leonardo Martinez Ortiz
Clustering tensor flow con kubernetes y raspberry pi
- 1. Clustering Tensor Flow con Kubernetesy Raspberry Pi Andres L Martinez @davilagrau Photo by William Felker on Unsplash
- 2. almo Google Developer ProgramLead PAN EU @davilagrau https://www.linkedin.com/in/aleonar https://www.instagram.com/davilagrau https://github.com/almo https://www.facebook.com/davilagrau
- 3. Lucas Käldström’s Motivation ●Kubernetes’ Dev Community ○ committer ○ maintainer ● Main motivations ○ Learning Google’s technologies ○ Developing Open Source ○ Re-use old/cheap HW
- 4. Photo by KenTreloar on Unsplash ● Hyperparameter tuning ○ Auto ML ● Scaling on QPS ○ ML API ● Ensemble learning ● Data Parallelism Model Parallelism? Don’t ask, don’t tell
- 5.
- 6.
- 7. Raspberry Pi 3 ●Single-board computer ● ARM 1.2 GHz 64/32-bit quad-core ○ VFPv4 Floating Point Unit onboard (per core) ○ Hardware virtualization support ● 1 GB LPDDR2 RAM at 900 MHz ● MicroSDHC slot ● Bluetooth 4.1 ● 2.4 GHz WiFi 802.11n & Ethernet 10/100
- 8.
- 9. Kubernetes ● Kubernetes isan open-source system for automating deployment, scaling, and management of containerized applications. ● Horizontal scaling ● Service discovery and load balancing ● Self-healing
- 10.
- 11. TensorFlow ● TensorFlow isan open source software library for numerical computation using data flow graphs. ● TensorFlow has APIs available in C++, Python, Java and Go. ● TensorFlow has also bindings for: C#, Haskell, Julia, Ruby, Rust, and Scala. ● TensorFlow Lite is TensorFlow’s lightweight solution for mobile and embedded devices Raspberry Pi version is coming soon!
- 12. Architecture HypriotOS HypriotOS HypriotOSHypriotOS Kubernetes Cluster Tensor Flow Cluster
- 13.
- 14. Setting Kubernetes up! Master Kubeadm Node#2 Node#1 Node#3 Kubelet Docker KubeletDocker Kubelet Docker
- 15. Setting up themaster Settings ● OS: Installing Docker on Raspbian OR ○ Download and flash HypriotOS v1.7.1 from https://goo.gl/y9Jyzd ● Setting up Kubernetes repositories ○ Key / Source Master Kubeadm Commands ● apt-get update && apt-get install -y kubeadm ● echo `cat /boot/cmdline.txt` cgroup_enable=cpuset > /boot/cmdline.txt ● swapoff -a #Note: Kubernetes 1.8 ● kubeadm init --pod-network-cidr 10.244.0.0/16
- 16. Setting up thenode (each) Settings ● OS: Installing Docker on Raspbian OR ○ Download and flash HypriotOS v1.7.1 from https://goo.gl/y9Jyzd ● Setting up Kubernetes repositories ○ Key / Source Commands ● apt-get update && apt-get install -y kubeadm ● echo `cat /boot/cmdline.txt` cgroup_enable=cpuset > /boot/cmdline.txt ● swapoff -a #Note: Kubernetes 1.8 ● kubeadm join --token=XXXXX Master-IP Node #2 Kubelet Docker
- 17. Setting the network Flannel:flannel is a virtual network that gives a subnet to each host for use with container runtimes “Platforms like Google's Kubernetes assume that each container (pod) has a unique, routable IP inside the cluster. The advantage of this model is that it reduces the complexity of doing port mapping”
- 18. Let’s scale with AnsibleScripts https://github.com/lahsivjar/kube-arm
- 19.
- 20. Parallelization Strategies Distributed TensorFlow Explicit(device block): TensorFlow will insert the appropriate data transfers between the jobs. with tf.device(“/cpu:0”): a = tf.Variable(3.0) b = tf.Variable(3.0) c = a * b Parallelization strategies: ● In-graph replication ● Between-graph replication ● Asynchronous training ● Synchronous training TensorFlow Serving It might be also a function
- 21. TensorFlow Cluster A TensorFlow"cluster" is a set of "tasks" that participate in the distributed execution of a TensorFlow graph. Steps: 1. Create a tf.train.ClusterSpec that describes all of the tasks in the cluster. This should be the same for each task. 2. Create a tf.train.Server, passing the tf.train.ClusterSpec to the constructor, and identifying the local task with a job name and task index.
- 22. TensorFlow Cluster Node #0 tf.train.ClusterSpec({ "worker":[ "worker0.example.com:2222", "worker1.example.com:2222" ], "ps": [ "ps0.example.com:2222", "ps1.example.com:2222" ]}) Worker Node #1 Worker Node #1 PS Node #0 PS
- 23. Setting up TensorFlowCluster I Node #1 PS Node #0 PS $ python trainer.py --ps_hosts=ps0.example.com:2222,ps1.example.com:2222 --worker_hosts=worker0.example.com:2222,worker1.example.com:2222 --job_name=ps --task_index=0 $ python trainer.py --ps_hosts=ps0.example.com:2222,ps1.example.com:2222 --worker_hosts=worker0.example.com:2222,worker1.example.com:2222 --job_name=ps --task_index=1 Example: Between-graph replication / Asynchronous training
- 24. Setting up TensorFlowCluster II Node #1 Worker Node #0 Worker $ python trainer.py --ps_hosts=ps0.example.com:2222,ps1.example.com:2222 --worker_hosts=worker0.example.com:2222,worker1.example.com:2222 --job_name=worker --task_index=0 $ python trainer.py --ps_hosts=ps0.example.com:2222,ps1.example.com:2222 --worker_hosts=worker0.example.com:2222,worker1.example.com:2222 --job_name=worker --task_index=1
- 25. Sharding Variables inMultiples Parameters Servers with tf.device(tf.train.replica_device_setter()) with tf.train.MonitoredTrainingSession(master=server.target, is_chief=(FLAGS.task_index == 0), checkpoint_dir="/tmp/train_logs", hooks=hooks) as mon_sess: Node #1 PS Node #2 PS Node #0 PS
- 26. Docker Image CPU Onlyi.e. Raspberry Pi ● TensorFlow 1.1 https://goo.gl/URUpko ● Official TensorFlow Lite for Raspberry Pi, Coming Soon! https://goo.gl/viqtuQ ● resin/rpi-raspbian + tensorflow 1.4 Coming soon!
- 27.
- 28. Pod Controllers: StatefulSets (PS & Workers) ● Manages the deployment and scaling of a set of Pods. ● Provides guarantees about the ordering and uniqueness of these Pods. ● StatefulSet manages Pods that are based on an identical container spec. ● StatefulSet maintains a sticky identity for each of their Pods. StatefulSet #0 Node #1 PS Node #0 PS Node #1 Worker Node #0 Worker StatefulSet #1
- 29.
- 31. Thank you! Questions? Andres LMartinez @davilagrau
- 32. Setting the clusterI $ kubectl run hypriot --image=hypriot/rpi-busybox-httpd --replicas=3 --port=80 $ kubectl expose deployment hypriot --port 80 $ kubectl get endpoints hypriot 1 2 3
- 33. Setting the clusterII $ kubectl apply -f traefik-ingress-controller.yaml $ kubectl label node IP nginx-controller=traefik $ kubectl apply -f cluster-ingress.yaml 1 2 3 $ cat cluster-ingress.yaml apiVersion: extensions/v1beta1 kind: Ingress metadata: name: hypriot spec: rules: - http: paths: - path: / backend: serviceName: hypriot servicePort: 80
- 34. Kubernetes: orquestación deimagenes de Rasberri Pi para el despliegue de tensorflow server. Description: Structure: - Use case: IoT / processing information - Architecture Kubernetes + Tensorflow+ rasberri pi - Python - Introduction to Kubernetes (Ansible?) (Laura) - Master / Slave architecture (Laura) - Container description: labelling and pod matching (Laura) - Configuration (Laura) - Load balancer (Laura) - round robin HTTP request (Laura) - Monitoring load of the replicas (Laura) - Failure tolerance (Laura) - Introduction to Tensor Flow - Introduction to TensorFlow / MachineLearning - Computation Graph - Introduction to TensorFlow Server - Development of Use Case
- 35. Dashboard codemotion-1 192.168.1.76 B8:27:EB:E5:9D:A5 codemotion-3192.168.1.75 B8:27:EB:68:5B:F9 codemotion-4 192.168.1.77 B8:27:EB:1F:EF:29 codemotion-2 192.168.1.78 B8:27:EB:19:38:C3
- 36. Laura Morillo-Velarde ● Backendengineer at seedtag ● Twitter: @Laura_Morillo ● WTM Lead at GDG Madrid