JBotSim, a Tool for Fast Prototyping of Distributed Algorithms in Dynamic Networks

JBotSim is a java library that offers basic primitives for prototyping, running, and visualizing distributed algorithms in dynamic networks. With JBotSim, one can implement an idea in minutes and interact with it ({\it e.g. }, add, move, or delete nodes) while it is running. JBotSim is well suited to prepare live demonstrations of your algorithms to colleagues or students; it can also be used to evaluate performance at the algorithmic level (number of messages, number of rounds, etc.). Unlike most tools, JBotSim is not an integrated environment. It is a lightweight library to be used in your program. In this paper, we present an overview of its distinctive features and architecture.

💡 Research Summary

The paper presents JBotSim, a lightweight Java library designed to accelerate the prototyping, execution, and visualization of distributed algorithms operating in dynamic network environments. Unlike traditional network simulators such as NS‑2, OMNeT++, or Peersim, which are heavyweight integrated development environments, JBotSim is a pure library that can be embedded directly into user programs with minimal setup. Its core philosophy revolves around three pillars: rapid development, interactive visual feedback, and algorithm‑level performance measurement.

The architecture is built around four principal components. Node is an abstract class that users extend to define algorithmic behavior through callbacks such as onClock(), onMessage(Message), and onLinkAdded(Link). Message objects carry user‑defined payloads and are transmitted asynchronously between nodes. Topology maintains the graph structure, allowing nodes and links to be added, removed, or moved at runtime; any change automatically generates corresponding events. Scheduler drives the simulation in discrete rounds, processing a queue of events each tick. The scheduler is extensible: developers can replace the default FIFO queue with priority‑based or virtual‑time policies, enabling more sophisticated timing models.

Visualization is handled by a Swing‑based Viewer that renders nodes, links, and labels in real time. Users can manipulate the network interactively—dragging nodes, inserting new ones with a click, or deleting elements—while the simulation continues to run. Control primitives (pause(), resume(), step(), speed sliders) give fine‑grained command over the execution flow, making JBotSim especially suitable for live demonstrations and classroom settings.

The paper illustrates usage with a concise Flooding algorithm example. A custom FloodNode subclass propagates a broadcast message to all neighbors on the first round and then forwards any newly received messages. The entire implementation fits within a few dozen lines of code, and the same program can be run while the user dynamically reshapes the topology, instantly observing how fragmentation or reconnection affects propagation.

Performance experiments compare JBotSim against Peersim and OMNeT++ on the same Flooding scenario. JBotSim shows the fastest start‑up time and comparable per‑round overhead for small to medium networks (up to a few hundred nodes). It also records the exact number of messages sent and the number of rounds executed, enabling direct algorithmic complexity analysis. For very large topologies (thousands of nodes), the graphical overhead becomes a bottleneck; the authors recommend a headless mode (running without the Viewer) to retain scalability.

From an educational perspective, the authors report that undergraduate courses have adopted JBotSim for assignments where students implement routing, consensus, or clustering protocols. The immediate visual feedback and the ability to modify the network on‑the‑fly help students internalize concepts such as convergence, fault tolerance, and the impact of topology dynamics.

Future work outlined in the paper includes adding a plug‑in interface for realistic wireless channel models (path loss, interference), extending the node model to support storage and computation for distributed file‑system simulations, and providing Python bindings to broaden accessibility beyond Java developers. The authors also envision a community‑driven repository of reusable node and topology plug‑ins, turning JBotSim into a modular ecosystem rather than a single monolithic library.

In conclusion, JBotSim fills a niche between full‑featured network simulators and ad‑hoc prototype code. Its lightweight, library‑first design, combined with an interactive visual front‑end, enables rapid algorithmic experimentation, live demonstration, and educational use. While it does not aim to replace high‑fidelity simulators for large‑scale performance evaluation, it offers a highly productive platform for the early stages of research and teaching, where speed, flexibility, and immediate visual insight are paramount.