← All stories
● Covered by 1 source · 1 reportLow impact1 neutral

Running 15 Million Web Servers on a Single VM Using smoltcp, DPDK, and AF_XDP

🔄 Updated 2d ago
New to BrevFeed? We gather this story from every outlet covering it into one summary — ranked by real-world impact, not just the latest headline — so you never miss what matters. What is BrevFeed? →

Key points

  • Goal: 15 million web servers on 1 VM (10GB RAM, 1 CPU).
  • Initial C server achieved 65k RPS but hit port limits.
  • Explores smoltcp, DPDK, and AF_XDP for network optimization.
  • Addresses challenges of high concurrency and resource usage.

The Challenge: Extreme Web Server Density

The article sets out to demonstrate the feasibility of running 15 million web servers on a single virtual machine. The specified constraints for this experiment are 10GB of RAM and one CPU core. This ambitious goal aims to explore the limits of network stack optimization and resource management for high-concurrency applications.

Initial Web Server Implementation and Limitations

A basic web server implemented in C, utilizing standard syscalls like listen, accept, recv, and send, was initially developed. This simple server achieved approximately 65,000 requests per second (RPS). However, when attempting to scale to millions of instances, the server quickly encountered limitations related to available network ports, specifically exhausting the kernel's ip_local_port_range.

Advanced Networking Techniques for Scale

To overcome the identified limitations and achieve the target of 15 million web servers, the article indicates the use of advanced networking technologies. These include smoltcp, a Rust-based TCP/IP stack, DPDK (Data Plane Development Kit) for high-performance packet processing, and AF_XDP (Address Family eXpress Data Path) for efficient kernel bypass networking. These tools are crucial for managing the high volume of connections and optimizing network throughput within the constrained environment.

Addressing Port Exhaustion and Resource Management

The core problem of port exhaustion when assigning a unique port to each web server is a central focus. The article implies that the chosen networking technologies will be instrumental in managing connections without relying solely on the kernel's ephemeral port range. This involves rethinking how network resources are allocated and handled at a low level to support millions of concurrent server instances efficiently.

✨ This summary was generated by AI from the outlets' reporting listed below. It is not independently verified and may contain errors — check the original sources. How BrevFeed works →

The daily brief

One email each morning: the day's tech stories, clustered across outlets and summarized. No account needed.

One email a day. Unsubscribe in one click, any time.

Today's brief

Spend a few minutes, get the whole day. Every topic's top stories in one hands-free rundown — listen, watch, or read the transcript.

~4 min · 3 stories · Nov 21

▶ Play today's brief Listen on Spotify

New every morning, and the back catalogue is archived by date.

Reporting from

This article explores the technical challenge of hosting 15 million web servers on a single virtual machine with limited resources (10GB RAM, 1 CPU core). It details the use of smoltcp, DPDK, and AF_XDP to achieve high-density web server deployment. The author demonstrates how to overcome port limitations and optimize network stack performance for extreme concurrency.