Linux Scalability Effort Homepage

Mission

To drive Single System Image scalability issues in Linux both from a large systems perspective (large processor counts, IO configs, etc) as well a from an intensive enterprise workload perspective (large number of processes, memory requirements, bandwidth requirements).

Introduction

This project is made up of several groups interested in using Linux on larger systems and to solve larger problems. Several hardware vendors have committed to Linux running on larger and larger classes of machines as well as with more and more intensive workloads. This project is a community based project for all developers interested increasing the scalability of Linux to meet these challenges. The Linux Scalability Effort is not a traditional SourceForge project. Rather, it is a collection of sub-projects and activities all centered around the common goal of scalability. Recent news regarding the project can be found here.

Sub-Projects/Activities

Project Resources

Project management tools and mailing lists can be found on the SourceForge site.

Communications

We host a regular conference call which is open to project members to report status and discuss general issues. Minutes are archived. We also host technical discussions on IRC. More info can be found here.

Tools

Workloads

In order to facilitate scalability studies, we need to identify exactly how we can claim an OS is "scalable" or "unscalable". Doing this can be quite difficult, specially since the OS might scale quite well under certain loads, and degrade quite heavily under other loads. Finally, what users care about is how the system scales with _their_ workloads. But, each set of users puts the system to different uses. While one system might be exclusively used as a web server, another might be used as an nfs server, yet another as a build server or compute server. To try to satisfy the scalability requirements of each specific user, or application category, we need to have different benchmarks to simulate loading the OS under those conditions. In some cases, it becomes too complex to study loading different parts of the OS (imagine studying memory loading along with disk elevator performance while swapping, or some such interacting subsystem analysis), so it might make more sense throwing simpler-to-understand, aka synthetic benchmarks at the system.

Yet another problem is the ability to talk about benchmark numbers in the open as certain benchmarks have specific restrictions on their use, usually requiring the results to be audited and approved. We have set up a sister project, the Linux Benchmark Suite, to collect different benchmarks for Linux to aid this project as well as others. You can check out results for several benchmarks here.

Roadmap

A somewhat dated project roadmap is available. See here for more information and alternative file formats.

Other useful resources

This page recovered from an archive on May 16, 2005. The date on the archive was June 28, 2004. Any problems, additions, etc , please send email to the Webmeistern.

SourceForge.net Logo