🚀 ContSysSim - Python based Containerized System Simulation

Container Simulation Framework is a Docker Swarm-like simulation tool built with SimPy. It allows users to simulate the behavior of Nodes and Containers, track resource utilization, and visualize system performance over time.

📖 Overview

This framework enables users to:

• Define Nodes (Physical or Virtual Machines) with limited resources (CPU, RAM, Disk, Bandwidth).
• Deploy multiple Containers inside Nodes with dynamic workloads.
• Apply fluctuation effects on resource consumption.
  • Description of fluctuation calculations
• Simulate workload requests with different delay, duration, and priority.
• Monitor and visualize CPU, RAM, Disk, and Bandwidth utilization.
• Use First-Fit Load Balancing to allocate workloads efficiently.
• Enable or disable resource reservations for smarter workload scheduling.
• Monitoring and data collection of an existing containerized system.
  • Description of container Analyzer tool

---

🏗️ High-Level Architecture

Below is the high-level architecture illustrating the framework's structure and its components:

---

🔧 Local Installation

1️⃣ Clone the Repository

git clone https://github.com/milanbalazs/ContSysSim
cd ContSysSim

2️⃣ Create a Virtual Environment (with container-simulation)

./tools/create_venv.sh

This script creates a Python virtual environment (simulation_venv), installs the required dependencies, and container-simulation package.

3️⃣ Activate the Virtual Environment

source simulation_venv/bin/activate

4️⃣ (Optional) Install the Package Locally

pip install -e .

This installs the package in editable mode, allowing development without reinstallation.

---

🚀 How to Use?

1️⃣ Run a Simple Simulation

A basic simulation with a single Node and one container can be executed using (simple.py):

python examples/simple.py

2️⃣ Run a Multi-Node Simulation

For a multi-node environment with multiple Nodes and containers (multi_node.py):

python examples/multi_node.py

3️⃣ Configuration-Based Simulation

The framework also supports running simulations entirely through a YAML configuration file. This allows you to define the entire simulation setup, including Nodes, Containers, Workloads, and Load Balancer configurations, without modifying the code.

Steps to Run

1. Create a YAML configuration file. See the Configuration Guide for details and examples.
2. Run the simulation by specifying the configuration file:

python cli_starter.py --config path/to/config.yml

Example Configuration

Refer to the Configuration Guide for a detailed breakdown of the configuration structure and an example YAML file.

---

4️⃣ Enable or Disable Load Balancer Reservations

With the First-Fit with Reservations Load Balancer, users can toggle workload reservations using the use_reservations parameter:

✅ First-Fit with Reservations (Default)

The scheduler considers workload start time and duration when assigning resources.

simulation = LbSimulation(use_reservations=True)

⚡ Classic First-Fit (No Reservations)

The scheduler ignores workload timing and places them immediately if resources are available.

simulation = LbSimulation(use_reservations=False)

---

🛠 Key Components

🔹 1. Node (Physical or Virtual machines)

• Represents a host with limited resources (CPU, RAM, Disk, Bandwidth).
• Hosts multiple containers and manages resource allocation.
• Monitors resource utilization over time.
• Supports fluctuation effects for realistic fluctuations.

🔹 2. Container

• Runs inside a Node and consumes computing resources dynamically.
• Can receive workload requests, increasing resource usage temporarily.
• Tracks historical usage of CPU, RAM, Disk, and Bandwidth.

🔹 3. Workload Requests

• Simulates incoming tasks that containers process.
• Each workload request has:
  • CPU, RAM, Disk, Bandwidth requirements.
  • Delay before execution starts.
  • Duration for which it runs.
  • fluctuation percentage to simulate fluctuating demand.

🔹 4. Load Balancer

• Implements First-Fit with Reservations Scheduling for efficient workload placement.
• Uses the use_reservations parameter to toggle between:
  • Smart scheduling (default) → Considers workload start time & duration.
  • Immediate scheduling → Places workloads as soon as resources are available.
• Prevents resource overload by rejecting excessive workloads.

🔹 5. Simulation Engine

• Uses SimPy to manage event-driven execution.
• Handles Node startup delays, container execution, and monitoring.
• Runs the simulation for a fixed duration and logs resource usage.

🔹 6. Visualization

• Uses Matplotlib to plot resource usage over time.
• Tracks and displays:
  • CPU, RAM, Disk, and Bandwidth consumption.
  • Available resources after workload execution.
  • Fluctuations over time.

---

📊 Example Output

After running a simulation, you will see terminal logs like:

[10] Node 'manager-1' Status - Containers: 3, CPU: 2.5/8.0 RAM: 2048/16384 Disk: 4096/20480 BW: 600/10000
[12] Container 'KeyCloak' updated workload: CPU 1.0/2.0 --> 1.5/2.0, RAM 1024/2048 --> 1536/2048

And generate graphs like:

📈 CPU, RAM, Disk, etc... usage over time

---

🎯 Contributing

Want to contribute? Follow these steps:

1. Fork the repo and clone it locally.
2. Create a new branch (git checkout -b feature-branch).
3. Make your changes and add docstrings.
4. Push your changes (git push origin feature-branch).
5. Create a pull request 🚀.

---

💡 Troubleshooting

Q: "ModuleNotFoundError: No module named 'cont_sys_sim'"

• Ensure you've activated the virtual environment (source simulation_venv/bin/activate).
• Reinstall the package: pip install -e .

Q: "Simulation hangs indefinitely"

• Ensure you set an appropriate simulation_time in your script.

---

📜 License

This project is licensed under the MIT License.

---

🙌 Acknowledgments

TODO

---

🔗 Author: Milan Balazs
🔗 GitHub Repo: Container Simulation Framework