> ## Documentation Index
> Fetch the complete documentation index at: https://cloudsim-ho-project.puneetchandna.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Architecture

> High-level overview of the project's architecture.

This page provides a high-level overview of the architecture of the CloudSim-HO-Research-V2 project.

## The Hippopotamus Optimization (HO) Algorithm

The core of this research framework is the Hippopotamus Optimization (HO) algorithm, a nature-inspired metaheuristic algorithm that models the social behavior of hippos to solve complex optimization problems.

### Core Concepts

* **Population:** A set of "hippos," where each hippo represents a potential solution to the VM placement problem.
* **Fitness:** A multi-objective function that evaluates the quality of each solution based on resource utilization, power consumption, and SLA violations.
* **Position Update:** The process by which hippos move in the search space, influenced by the "leader" (the best solution) and "prey" (random solutions).

### Implementation Details

The HO algorithm is implemented in the `org.puneet.cloudsimplus.hiippo.algorithm` package, with the following key classes:

* **`HippopotamusOptimization.java`:** The main class that orchestrates the optimization process.
* **`Hippopotamus.java`:** Represents a single solution (a hippo) in the population.
* **`HippopotamusParameters.java`:** Encapsulates all the tunable parameters for the HO algorithm.

## Simulation Workflow

The simulation is coordinated by the `ExperimentCoordinator.java` class, which manages the entire experimental workflow.

### The Workflow Steps

1. **Configuration:** The coordinator loads the experimental setup from `config.properties`, including the scenarios and algorithms to be tested.
2. **Scenario Generation:** For each scenario, the `TestScenarios.java` class creates a set of VMs and hosts with specific configurations.
3. **Experiment Execution:** The `ExperimentRunner.java` class runs the simulation for each algorithm and scenario, for a specified number of replications.
4. **Metrics Collection:** During the simulation, the `MetricsCollector.java` class gathers all the relevant performance data.
5. **Results Writing:** The `CSVResultsWriter.java` class writes all the collected data to CSV files in the `results/` directory.
6. **Statistical Analysis:** Finally, the `ComparisonAnalyzer.java` class performs statistical tests to compare the performance of the different algorithms.