1. Introduction

1.1. What is Tsung?

Tsung (formerly IDX-Tsunami) is a distributed load testing tool. It is protocol-independent and can currently be used to stress HTTP, WebDAV, SOAP, PostgreSQL, MySQL, AMQP, MQTT, LDAP and Jabber/XMPP servers.

It is distributed under the GNU General Public License version 2.

1.2. What is Erlang and why is it important for Tsung?

Tsung’s main strength is its ability to simulate a huge number of simultaneous user from a single machine; moreover, you can distribute the users on cluster for machines. When used on cluster, you can generate a really impressive load on a server with a modest cluster, easy to set-up and to maintain. You can also use Tsung on a cloud like EC2.

Tsung is developed in Erlang and this is where the power of Tsung resides.

Erlang is a concurrency-oriented programming language. Tsung is based on the Erlang OTP (Open Telecom Platform) and inherits several characteristics from Erlang:

Erlang has been made to support hundred thousands of lightweight processes in a single virtual machine.
Erlang runtime environment is naturally distributed, promoting the idea of process’s location transparency.
Erlang has been built to develop robust, fault-tolerant systems. As such, wrong answer sent from the server to Tsung does not make the whole running benchmark crash.

More information on Erlang on http://www.erlang.org.

1.3. Tsung background


  • Tsung development was started by Nicolas Niclausse in 2001 as a distributed jabber load stress tool for internal use at http://IDEALX.com/ (now OpenTrust). It has evolved as an open-source multi-protocol load testing tool several months later. The HTTP support was added in 2003, and this tool has been used for several industrial projects. It is now hosted on github, and several companies provide profesionnal support. The list of contributors is available in the source archive at https://github.com/processone/tsung/blob/master/CONTRIBUTORS.

  • It is an industrial strength implementation of a stochastic model for real users simulation. User events distribution is based on a Poisson Process. More information on this topic in:

    Z. Liu, N. Niclausse, and C. Jalpa-Villanueva. Traffic Model and Performance Evaluation of Web Servers. Performance Evaluation, Volume 46, Issue 2-3, October 2001.

  • This model has already been tested in the INRIA WAGON research prototype (Web trAffic GeneratOr and beNchmark). WAGON was used in the http://www.vthd.org/ project (Very High Broadband IP/WDM test platform for new generation Internet applications, 2000-2004).

Tsung has been used for very high load tests:

  • Jabber/XMPP protocol:
    • 90,000 simultaneous Jabber users on a 4-node Tsung cluster (3xSun V240 + 1 Sun V440).
    • 10,000 simultaneous users. Tsung was running on a 3-computers cluster (CPU 800MHz).
    • 2,000,000 concurrent users on a single m4.10xlarge instance on EC2 to tests ejabberd scalability
  • HTTP and HTTPS protocol:
    • 12,000 simultaneous users. Tsung were running on a 4-computers cluster (in 2003). The tested platform reached 3,000 https requests per second.
    • 10 million simultaneous users running on a 75-computers cluster, generating more than one million requests per second.

Tsung has been used at:

  • DGI (Direction Générale des impôts): French finance ministry
  • Cap Gemini Ernst & Young
  • IFP (Institut Français du Pétrole): French Research Organization for Petroleum
  • LibertySurf
  • Sun (TM) for their Moodlerooms platform on Niagara processors: https://blogs.oracle.com/kevinr/resource/Moodle-Sun-RA.pdf
  • and many other companies