Lapse

This repository contains the source code and the dataset used in the paper Lapse: Latency & Power-Aware Placement of Data Stream Applications on Edge Computing. Lapse is a cost-based heuristic algorithm for the placement of stream processing applications on edge computing environments.

Installation and usage guide

This section provides instructions on setting up your environment for running simulations and replicating the paper's results.

Prerequisites

Before you can execute the simulations, ensure that you have the following packages installed:

• Python 3.9
• Poetry 1.7.1

To install the required dependencies, follow these steps:

poetry install
poetry shell

Reproducing the results

To replicate the paper's results, you can either run the following commands or execute the Jupyter notebook analysis.ipynb.

First, optionally create the dataset:

python dataset.py --name dataset > datasets/dataset.log

Next, run the simulations for the algorithms:

python -B -m simulator --dataset datasets/dataset.json --algorithm storm
python -B -m simulator --dataset datasets/dataset.json --algorithm storm_la
python -B -m simulator --dataset datasets/dataset.json --algorithm aels
python -B -m simulator --dataset datasets/dataset.json --algorithm aels_pa
python -B -m simulator --dataset datasets/dataset.json --algorithm lapse

Manuscript

Available soon