Compare commits
11 Commits
v2.0.0-bet
...
sample-run
| Author | SHA1 | Date | |
|---|---|---|---|
| e6bd7ae103 | |||
| 5f76b1272a | |||
| 15b3344238 | |||
| 1324e0394f | |||
| 6c86082f48 | |||
|
|
5a164749b3 | ||
|
|
31b9e7e934 | ||
| 9a72149a52 | |||
| 3e1e6e6928 | |||
| 9809517ebd | |||
| 7742377304 |
7
.gitignore
vendored
7
.gitignore
vendored
@@ -7,4 +7,9 @@ server/docker/__pycache__
|
||||
p2p-rendering-computation
|
||||
p2prc
|
||||
config.json
|
||||
.vscode/
|
||||
*.tgz
|
||||
.vscode/
|
||||
hosts
|
||||
*.retry
|
||||
**/*.swp
|
||||
**/*.retry
|
||||
|
||||
2
Makefile
2
Makefile
@@ -1,4 +1,4 @@
|
||||
SHELL := /bin/bash
|
||||
/ https://www.youtube.com/channel/UCQdZltW7bh1ta-_nCH7LWYw/playlistSHELL := /bin/bash
|
||||
|
||||
install:
|
||||
go build -o p2prc
|
||||
|
||||
@@ -13,5 +13,7 @@ general:
|
||||
- software-properties-common=0.96.24.32.7 # known good with 0.96.24.32.7
|
||||
- git
|
||||
- golang
|
||||
- g++
|
||||
- openjdk-8-jdk
|
||||
pip:
|
||||
- docker>=3.7.0 # known good with 3.7.0
|
||||
|
||||
2
go.mod
2
go.mod
@@ -29,7 +29,7 @@ require (
|
||||
golang.org/x/sys v0.0.0-20210320140829-1e4c9ba3b0c4 // indirect
|
||||
golang.org/x/text v0.3.5 // indirect
|
||||
golang.org/x/time v0.0.0-20210220033141-f8bda1e9f3ba // indirect
|
||||
golang.org/x/tools v0.0.0-20210106214847-113979e3529a
|
||||
golang.org/x/tools v0.0.0-20210106214847-113979e3529a // indirect
|
||||
google.golang.org/grpc v1.35.0 // indirect
|
||||
gopkg.in/yaml.v2 v2.4.0 // indirect
|
||||
gotest.tools/v3 v3.0.3 // indirect
|
||||
|
||||
@@ -5,7 +5,7 @@
|
||||
# 3. handlers - these steps run only in response to specific events. If the event does not occur, the handler does not run.
|
||||
################################################################################
|
||||
---
|
||||
- hosts: test # Alternatively, 'canary', 'blue'|'green', 'all' according to your deployment strategy and your /etc/ansible/hosts file
|
||||
- hosts: gpu # Alternatively, 'canary', 'blue'|'green', 'all' according to your deployment strategy and your /etc/ansible/hosts file
|
||||
become: yes # All commands must run as root
|
||||
strategy: free # Allows all hosts to run to the end of the playbook as fast as they can
|
||||
|
||||
@@ -26,8 +26,8 @@
|
||||
# vars:
|
||||
# users: "{{ secrets.users }}"
|
||||
|
||||
- role: cuda # CUDA Driver Setup
|
||||
tags: [cuda]
|
||||
# - role: cuda # CUDA Driver Setup
|
||||
# tags: [cuda]
|
||||
|
||||
# - role: nvidia # NVIDIA-Docker Runtime Setup
|
||||
# tags: [nvidia]
|
||||
3
sample-run/ML/requirements.yml
Normal file
3
sample-run/ML/requirements.yml
Normal file
@@ -0,0 +1,3 @@
|
||||
---
|
||||
- name: nvidia.nvidia_driver
|
||||
version: 2.0.0
|
||||
@@ -11,5 +11,10 @@ general:
|
||||
- software-properties-common # known good with 0.96.24.32.7
|
||||
- git
|
||||
- golang
|
||||
- apt-utils
|
||||
- libopenmpi-dev
|
||||
- python3-setuptools
|
||||
- python3-dev
|
||||
pip:
|
||||
- docker>=3.7.0 # known good with 3.7.0
|
||||
- tensorflow
|
||||
27
sample-run/run.txt
Normal file
27
sample-run/run.txt
Normal file
@@ -0,0 +1,27 @@
|
||||
Create 3 vm using ./p2prc --CreateVM 0.0.0.0 --Ports=1 --GPU
|
||||
|
||||
then on each run
|
||||
|
||||
sudo apt clean && sudo apt update && sudo apt -y install cmake ccache nano
|
||||
sudo ln -s /usr/local/cuda-11.2 /usr/local/cuda
|
||||
export PATH=${HOME}/.local/bin:$PATH && export PATH=/usr/local/cuda-11.2/bin${PATH:+:${PATH}} && export LD_LIBRARY_PATH=/usr/local/cuda-11.2/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
|
||||
wget -c https://raw.githubusercontent.com/horovod/horovod/master/examples/tensorflow2/tensorflow2_synthetic_benchmark.py
|
||||
|
||||
pip install --no-cache-dir horovod[tensorflow]
|
||||
|
||||
# this will test if its locally install properly
|
||||
horovodrun -np 1 python3 tensorflow2_synthetic_benchmark.py
|
||||
|
||||
then configure machine1 ssh/config file like
|
||||
Host host1
|
||||
HostName 0.0.0.0
|
||||
Port 39751
|
||||
|
||||
Host host2
|
||||
HostName 0.0.0.0
|
||||
Port 34779
|
||||
|
||||
and ensure that the ssh can be connected and ssh keys are copied using ssh-copy-id command
|
||||
|
||||
then run
|
||||
horovodrun -np 2 -H host1:1,host2:1 python3 tensorflow2_synthetic_benchmark.py
|
||||
@@ -1,12 +0,0 @@
|
||||
---
|
||||
all:
|
||||
vars:
|
||||
ansible_python_interpreter: /usr/bin/python3
|
||||
test:
|
||||
hosts:
|
||||
test1:
|
||||
ansible_host: 0.0.0.0 # Replace with your remote IP
|
||||
ansible_port: 44003 # Replace with your remote SSH port
|
||||
ansible_user: master # Replace wtih your username
|
||||
ansible_ssh_pass: password
|
||||
ansible_sudo_pass: password
|
||||
@@ -1,23 +0,0 @@
|
||||
# The `cuda` Role
|
||||
|
||||
The `cuda` role installs the Nvidia CUDA GPU repository and drivers.
|
||||
|
||||
This role may be executed independently by running:
|
||||
|
||||
```bash
|
||||
$ ansible-playbook packages.yml --tags "cuda"
|
||||
```
|
||||
|
||||
On execution, this role installs the following directly to the remote machine's operating system:
|
||||
|
||||
### Apt Repositories (Debian/Ubuntu)
|
||||
|
||||
- Nvidia CUDA
|
||||
|
||||
### Apt GPG Signing Keys (Debian/Ubuntu)
|
||||
|
||||
- cudatools <cudatools@nvidia.com>
|
||||
|
||||
### Operating System Packages (Debian/Ubuntu)
|
||||
|
||||
- cuda v10.x
|
||||
Binary file not shown.
Binary file not shown.
@@ -1,5 +0,0 @@
|
||||
---
|
||||
# CUDA Driver Setup
|
||||
- name: reboot the machine
|
||||
reboot:
|
||||
listen: "reboot the machine"
|
||||
@@ -1,22 +0,0 @@
|
||||
---
|
||||
# Cuda Driver Package Installation
|
||||
- name: copy the CUDA repo onto the machine
|
||||
copy:
|
||||
src: "{{ cuda.apt.repo }}"
|
||||
dest: "{{ general.working_dir }}/{{ cuda.apt.repo }}"
|
||||
|
||||
- name: get the CUDA signing key
|
||||
apt_key:
|
||||
state: present
|
||||
url: "{{ cuda.apt.signing_key_url }}"
|
||||
|
||||
- name: add the CUDA repo
|
||||
apt:
|
||||
deb: "{{ general.working_dir }}/{{ cuda.apt.repo }}"
|
||||
state: present
|
||||
|
||||
- name: install the CUDA drivers
|
||||
apt:
|
||||
name: "{{ cuda.apt.package }}"
|
||||
update_cache: yes
|
||||
notify: "reboot the machine"
|
||||
@@ -1,10 +0,0 @@
|
||||
---
|
||||
cuda:
|
||||
apt:
|
||||
package: cuda=11.* # known good with 10.0.130-1
|
||||
repo: cuda-11-3_11.3.1-1_amd64.deb # Network install
|
||||
dependency: cuda-runtime-11-3_11.3.1-1_amd64.deb
|
||||
signing_key_url: https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub
|
||||
|
||||
general:
|
||||
working_dir: /tmp
|
||||
@@ -1,3 +0,0 @@
|
||||
---
|
||||
- import_tasks: debian.yml
|
||||
when: ansible_facts['os_family']|lower == 'debian' # debian, ubuntu
|
||||
50
server/docker/containers/docker-ubuntu-sshd-gpu/Dockerfile
Normal file
50
server/docker/containers/docker-ubuntu-sshd-gpu/Dockerfile
Normal file
@@ -0,0 +1,50 @@
|
||||
# -----------------------------------------------------------------------------
|
||||
# This is base image of Ubuntu LTS with SSHD service.
|
||||
#
|
||||
# Authors: Art567
|
||||
# Updated: Sep 20th, 2015
|
||||
# Require: Docker (http://www.docker.io/)
|
||||
# -----------------------------------------------------------------------------
|
||||
|
||||
# Base system is the latest LTS version of Ubuntu.
|
||||
# from consol/ubuntu-xfce-vnc
|
||||
|
||||
# due to dependency issues vnc is still work in progress
|
||||
from nvidia/cuda:11.2.2-cudnn8-devel-ubuntu20.04
|
||||
|
||||
# Switch to root user to install additional software
|
||||
USER 0
|
||||
|
||||
# Make sure we don't get notifications we can't answer during building.
|
||||
env DEBIAN_FRONTEND noninteractive
|
||||
|
||||
# Prepare scripts and configs
|
||||
add ./scripts/start /start
|
||||
|
||||
# Download and install everything from the repos.
|
||||
run apt-get -q -y update; apt-get -q -y upgrade && \
|
||||
apt-get -q -y install sudo openssh-server python3-pip && \
|
||||
mkdir /var/run/sshd
|
||||
|
||||
# Set root password
|
||||
run echo 'root:password' >> /root/passwdfile
|
||||
|
||||
# Create user and it's password
|
||||
run useradd -m -G sudo master && \
|
||||
echo 'master:password' >> /root/passwdfile
|
||||
|
||||
# Apply root password
|
||||
run chpasswd -c SHA512 < /root/passwdfile && \
|
||||
rm /root/passwdfile
|
||||
|
||||
# Port 22 is used for ssh
|
||||
expose 22
|
||||
|
||||
# Assign /data as static volume.
|
||||
volume ["/data"]
|
||||
|
||||
# Fix all permissions
|
||||
run chmod +x /start
|
||||
|
||||
# Starting sshd
|
||||
cmd ["/start"]
|
||||
72
server/docker/containers/docker-ubuntu-sshd-gpu/README.md
Normal file
72
server/docker/containers/docker-ubuntu-sshd-gpu/README.md
Normal file
@@ -0,0 +1,72 @@
|
||||
Ubuntu LTS with SSH (Docker)
|
||||
=========
|
||||
|
||||
A Docker file to build a [docker.io] base container with Ubuntu LTS and a sshd service
|
||||
[docker.io]: http://docker.io
|
||||
Nice and easy way to get any server up and running using docker
|
||||
|
||||
|
||||
Instructions
|
||||
-----------
|
||||
- Install Docker first.
|
||||
Read [installation instructions] (http://docs.docker.io/en/latest/installation/).
|
||||
|
||||
|
||||
- Clone this repository:
|
||||
|
||||
`$ git clone https://github.com/art567/docker-ubuntu-sshd.git`
|
||||
|
||||
|
||||
- Enter local directory:
|
||||
|
||||
`$ cd docker-ubuntu-sshd`
|
||||
|
||||
- Change passwords in Dockerfile:
|
||||
|
||||
`$ vi Dockerfile`
|
||||
|
||||
- Build the container:
|
||||
|
||||
`$ sudo docker build -t art567/ubuntu .`
|
||||
|
||||
|
||||
- Run the container:
|
||||
|
||||
`$ sudo docker run -d=true --name=ubuntu --restart=always -p=2222:22 -v=/opt/data:/data art567/ubuntu /start`
|
||||
|
||||
|
||||
- Your container will start and bind ssh to 2222 port.
|
||||
|
||||
|
||||
- Find your local IP Address:
|
||||
|
||||
`$ ifconfig`
|
||||
|
||||
|
||||
- Connect to the SSH server:
|
||||
|
||||
`$ ssh root@<ip-address> -p 2222`
|
||||
|
||||
|
||||
- If you don't want to deal with root:
|
||||
|
||||
`$ ssh master@<ip-address> -p 2222`
|
||||
|
||||
|
||||
**VERY IMPORTANT!!!**
|
||||
-----------
|
||||
|
||||
**Don't forget to change passwords in Dockerfile before building image!**
|
||||
|
||||
|
||||
Versions
|
||||
-----------
|
||||
Some branches represents Ubuntu versions.
|
||||
|
||||
Branch `master` is always represented by latest Ubuntu LTS
|
||||
|
||||
For example:
|
||||
- 12.04 - Ubuntu 12.04 LTS
|
||||
- 14.04 - Ubuntu 14.04 LTS
|
||||
- 16.04 - Ubuntu 16.04 LTS
|
||||
|
||||
@@ -0,0 +1,11 @@
|
||||
#!/bin/bash
|
||||
# -----------------------------------------------------------------------------
|
||||
# docker-ubuntu-sshd /start script
|
||||
#
|
||||
# Authors: Art567
|
||||
# Updated: Sep 20th, 2015
|
||||
# -----------------------------------------------------------------------------
|
||||
|
||||
|
||||
# Run OpenSSH server in daemon mode
|
||||
/usr/sbin/sshd -D
|
||||
@@ -82,6 +82,9 @@ func BuildRunContainer(NumPorts int, GPU string) (*DockerVM,error) {
|
||||
return nil,err
|
||||
}
|
||||
RespDocker.ImagePath = config.DockerFile
|
||||
//if GPU == "true" {
|
||||
// RespDocker.ImagePath = "/home/asleepyguy/p2p-rendering-computation/server/docker/containers/docker-ubuntu-sshd-gpu/"
|
||||
//}
|
||||
|
||||
// Gets docker information from env variables
|
||||
cli, err := client.NewClientWithOpts(client.FromEnv, client.WithAPIVersionNegotiation())
|
||||
|
||||
0
server/docker/kill-containers.sh
Normal file → Executable file
0
server/docker/kill-containers.sh
Normal file → Executable file
Reference in New Issue
Block a user