logo
Kurashizu Blogwhere ideas flow
  • Home
  • News
  • Blog
  • About
  • Home
  • News
  • Blog
  • About

© 2026 Kurashizu. All rights reserved.

Admin·New Post·Service Status
Powered by Cloudflare·01.11.26
← Back to news

Co-Opting Linux Processes for High-Performance Network Simulation (2022)

Original article
usenix.org|17 points|0 comments|by teleforce|Jul 24, 2026

High-Performance Network Simulation via Linux Process Co-Option (2022)

Presented at the USENIX Annual Technical Conference (ATC 22), this research addresses the critical need for robust tools to develop, test, and evaluate distributed systems.

👥 Research Team

AuthorAffiliation
Rob JansenU.S. Naval Research Laboratory
Jim NewsomeTor Project
Ryan WailsGeorgetown University, U.S.

🔍 The Problem Statement

The authors argue that current network experimentation tools suffer from a binary failure:

"The state-of-the-art simulation tools are either prohibitively inefficient at large scales or are limited by nontrivial architectural challenges, inhibiting their widespread adoption."

Essentially, researchers are forced to choose between extreme inefficiency and architectural rigidity.

🚀 The Solution: Phantom

To bridge this gap, the team developed Phantom, a groundbreaking tool for distributed system experiments. Unlike traditional simulators, Phantom is a discrete-event network simulator that manages unmodified applications by treating them as standard Linux processes.

How it Works

Phantom "co-opts" these processes using a sophisticated synthesis of:

  1. Process Control: Managing the execution flow of the application.
  2. System Call Interposition: Intercepting syscalls to redirect network traffic.
  3. Data Transfer Methods: Ensuring efficient movement of information within the simulation.

📊 Performance Evaluation

The efficiency of Phantom was tested against several industry benchmarks. The results indicate significant speed improvements, particularly in large Peer-to-Peer (P2P) environments.

Speedup Comparison (P2P Benchmarks):

  • vs. Shadow: ≈2.2×\approx 2.2\times≈2.2× faster
  • vs. NS-3: ≈3.4×\approx 3.4\times≈3.4× faster
  • vs. gRaIL: ≈43×\approx 43\times≈43× faster

Note: In large-scale Tor network simulations, Phantom's performance remains comparable to that of Shadow.


🔓 Open Access & Availability

USENIX maintains a strict commitment to the open dissemination of research. Their policy includes:

  • Free access to all papers and proceedings upon event commencement.
  • Open availability of presentation slides.
  • Free access to all posted audio and video recordings.

📚 Citation Data

For those wishing to reference this work, use the following BibTeX entry:

@inproceedings {280766, 
  author = {Rob Jansen and Jim Newsome and Ryan Wails}, 
  title = {Co-opting Linux Processes for {High-Performance} Network Simulation}, 
  booktitle = {2022 USENIX Annual Technical Conference (USENIX ATC 22)}, 
  year = {2022}, 
  isbn = {978-1-939133-29-52}, 
  address = {Carlsbad, CA}, 
  pages = {327--350}, 
  url = {https://www.usenix.org/conference/atc22/presentation/jansen}, 
  publisher = {USENIX Association}, 
  month = jul 
}

🖼️ Associated Media

Below are the official badges and logos associated with the publication: