Overview: Using the ByteBlower Server
Posted by Pieter Vandercammen, Last modified by Craig Godbold on 19 January 2022 09:37 AM
In this article we will discuss the ByteBlower components from the server itself to the client computer. The main goal is to provide an overview of how they fit together and how they enable you to run traffic tests.
As shown in the picture above, the ByteBlower server performs the heavy-lifting to generate traffic and measurement results. This part is shown on the left. All systems are available as a 1U rack system. The client computer (shown left) drives the setup, even for traffic generation.
The ByteBlower Server chassis is host to 3 isolated systems:
The traffic generation software is created at Excentis. It takes full control of the Traffic Network interfaces and most of the hardware processing power. This software is regularly updated with new features and bugfixes. Even for these updates the ByteBlower server itself doesn’t require access to the internet which can be seen in the following article.
The host operating system fills in the general house-keeping. It takes control of the remaining network interfaces (MAN0, MAN1) and offers access through them. Other login possibilities are with a USB keyboard and display. Like the traffic generation software, this OS is updated together with the traffic generation software. No additional maintenance is required for this OS. Configuration is preferably done using the byteblower-configurator (see here for details).
The IPMI software is the third element of the ByteBlower server. This subsystem provides remote control to the ByteBlower server hardware. It can be used to boot-up, powerdown or inspect hardware issues. Unlike the previous two, this system is optional. For more information about the IPMI interface please refer to this article.
Client software is freely available for all major platforms. Testing is possible using the graphical ByteBlower GUI application or the command-line with the ByteBlower-CLT. Both share the same file formats, all stored locally on the client computer. These applications include a reporting engine that generates general reports in various formats, including HTML and PDF.
More scripting is available through Python and Tcl libraries, these allow you to use the ByteBlower API directly. The base building blocks are the same as found in the ByteBlower GUI but without the separate reporting engine. Overall, both programming languages give more flexibility for more complex scenarios and more results for in-depth analysis.
There are no software restrictions on the number of clients. In fact it's common to share a single system across multiple users. Download links are found in the following website:
Managing the ByteBlower server is possible using SSH. For the very first setup it's advised to connect a screen and keyboard to the system. For those familiar, non-physical access is also possible using IPMI.