distributed-election

Distributed election algorithms. Project for the exam "Sistemi Distribuiti e Cloud Computing".

Overview

This application, made in the context of the "Sistemi distribuiti e Cloud Computing" (9 CFU = 9 ETCS) course, followed at University of Rome Tor Vergata, aims to implement two distributed election algorithms, i.e.:

• H. Garcia-Molina's bully algorithm (1982)
• G. N. Fredrickson and N. A. Lynch algorithm, based on a ring topology (1987)

The project's functionalities are managed by some nodes, that elect a coordinator between them, and a service registry, that manages the node's network addresses.

Further information about this project are available into the docs folder.

Quickstart

You can either launch a local run or see the deployment in action.

Local Run

Some ready-to-use scripts are available into the examples folder. Follow these instructions to run them:

0. Install Go by following these instructions, then install gRPC and protocol-buffer if you don't have it in your machine:

go install google.golang.org/protobuf/cmd/protoc-gen-go@latest
go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@latest

curl -LO https://github.com/protocolbuffers/protobuf/releases/download/v3.15.8/protoc-3.15.8-linux-x86_64.zip
unzip protoc-3.15.8-linux-x86_64.zip -d $HOME/.local
export PATH="$PATH:$HOME/.local/bin"

1. Download the repository and apply privileges:

git clone https://github.com/massimostanzione/distributed-election
sudo chmod -R 770 distributed-election
cd distributed-election

2. Build the binaries for nodes and service registry:

cd scripts
./setup.sh

(Please run ./teardown.sh if something goes wrong.)

3. Just launch the algorithm you want:

cd ../examples

# to run a bully algorithm demo:
./bully.sh

# to run a Fredrickson-Lynch algorithm demo:
./fredricksonlynch.sh

... then just follow the instructions and enjoy!

Deployment

Notice: to execute this you have to have an active Amazon EC2 instance.

cd ../deployments/ansible

# use vim, nano, or whatever you prefer
vim hosts.ini

Substitute the following items:

• Replace HOST-IP-ADDRESS with your Amazon EC2 instance outbound IP address;
• Replace AMAZON_EC2_USERNAME with your Amazon EC2 instance username (default: ec2-user);
• Replace PATH/TO/KEYS.pem with your .pem file containing the private key you use to access your Amazon EC2 instance.

Now just deploy:

# install ansible, if you don't have it in your machine:
python3 -m pip install --user ansible

# start the deployment
ansible-playbook -v -i hosts.ini deploy.yaml

(if it is not your first Ansible run, the flag --flush-cache may be useful)

Now, to see what is running, connect to your Amazon EC2 instance:

ssh -i "PATH/TO/KEYS.pem" AMAZON_EC2_USERNAME@HOST-IP-ADDRESS

# Verify if containers are up and running
# ***NOTICE:*** Containers can take several (about 3, empirically) minutes before
# being made up and running! Be patient...
docker ps

# Inspect logs for all the containers
cd distributed-election/deployments
docker container logs

# or, to see log just for one container, take its name from 'docker ps' and run, from any position:
docker container logs CONTAINER_NAME

Keep it calm

If you already enjoyed the quickstart, you can now launch a more reasoned execution:

Assuming you already followed the setup section, move to the distributed-election folder and run:

cd bin

here you can find the executable node and serviceregistry. You can run them in separate terminal windows or place them into different nodes of a network and just run with

./node

or

./serviceregistry

Please take a look at the --help flag to see all the possible customization (further more of them, with the whole set of parameters that can be set, can be found into the configs folder.