VolPE

Volunteer Computing Platform for Evolutionary Algorithms

This repository houses the primary components of VolPE, i.e. the master and the worker applications.

Dependencies

1. Install podman and the Go toolchain
2. Setup podman appropriately
3. Download this repository
4. Export the environment variable CONTAINER_HOST. Value can be determined by running podman info -f json | jq .host.remoteSocket.path. The output from the command must be modified as follows.If the output is "/run/user/1000/podman/podman.sock", the following command must be used to export the env. var.

export CONTAINER_HOST=unix:///xyz.sock

Master Application

The following are instructions to get the master application up and running in different platforms:

Linux / MacOS / Windows:

1. cd into the master/ folder and run go run .

Worker Application

The following are instructions to get the worker application up and running in different platforms:

Linux / MacOS / Windows:

1. Set the VOLPE_MASTER environment variable to the IP address where the master is accessible, with port 8000 by default. For example, if the master is accessible at IP address 192.168.0.2:

export VOLPE_MASTER=192.168.0.2:8080

2. cd into the worker/ folder and run go run .

Steps to Fix (WSL)

To change the firewall configs in WSL do

mkdir -p ~/.config/containers nano ~/.config/containers/containers.conf

And add this line

[network] firewall_driver = "iptables"

NOTE Shutdown the WSL. And reboot. Only then the changes reflect

wsl --shutdown # Shutdown wsl wsl -d <distro-name> # Reboot

Building and Deploying a Problem

1. Clone repo https://github.com/aadit-n3rdy/volpe-py.git
2. cd into volpe-py/
3. Run bash build-image.sh
4. Ensure .tar file present

Running in the Website

1. Open browser and go to localhost:8000/static
2. Give the problem a name and choose .tar file built from volke-py/ and enter the memory count and number of target instances
3. Enter the problem name and click START PROBLEM
4. Enter the problem name and and click CONNECT STREAM