Qosium

Qosium is software for passive Quality of Service (QoS) and Quality of Experience (QoE) real-time performance measurement and monitoring. No matter whether you use wired or wireless networks, Qosium lets you discover the network performance from the perspective of real applications and services.


Qosium brochure

Qosiumin technical brochure

Passive QoS Measurement

Qosium takes you to a new level in measurement and analysis of network performance. The measurements are passive, which does not cause overhead to networks. Passive measurement means that the real existing application traffic is measured, instead of generating test traffic to be measured to the network. This reveals the true performance of currently active applications using the network.

Qosium is a solution for network administrators, network testing actors, application and solution developers, and device manufacturers. The same solution fulfills even the strictest requirements of different industry and logistics fields. For example, it enables monitoring and confirming that SLA agreements in terms of network performance are met. Industry robots and IoT devices can be connected reliably and with quality that satisfies their applications. If your business is to evaluate network performance, Qosium allows you to verify the performance with actual applications used without the need for generating artificial traffic for measuring.

Qosium helps network administrators understand and visualize network performance, which facilitates locating bottlenecks and problem areas in networks. Qosium's output can also be used to trigger, for example, different kinds of quality-driven alerts and actions. Network devices can be made more intelligent and end-user friendly. For example, in wireless mobility management, why rely only on signal strength criterion, which does not reflect application quality well, as the real application quality can also be made available.

How It Works

Qosium is a distributed measurement solution, which is composed of three types of components: the measurement agent, measurement controllers, and results systems. From these components, the measurement agent and a measurement controller are always needed in a Qosium measurement, while a results system is typically involved only in larger installations.

The measurement agent

Qosium Probe is a light-weight measurement agent, which is installed to network devices from which QoS is desired to be measured. Network devices can practically be anything that allows installing new software, including application servers, routers, switches, mobile devices, and IoT devices. Measurement can be carried out at a single point. However, all the features of Qosium come into use when measurement is between multiple points. Multipoint measurement enables measuring QoS performance end-to-end and over any part on the path.

Measurement controllers

A Qosium measurement is initiated with a measurement controller that connects to the QMCP interface of Qosium Probes. A measurement controller locates Qosium Probes in the network, parameterizes them, controls the measurement, and guides the results to be delivered where desired. It can be a pure controller, in which case the results are processed elsewhere, or an analyzer, being capable of analyzing and visualizing the results in addition to the measurement control. Kaitotek offers three measurement controllers, and also an opportunity to develop a custom controller.

Qosium Scope is an analyzer for creating, controlling, and parameterizing Qosium measurements. In addition, Scope collects and visualizes results. All the features of Qosium are available on Qosium Scope.

Qosium Scope Lite is a command-line tool for controlling measurements. It includes also limited real-time visualization capabilities. Scope Lite is especially suitable for constrained systems and creating measurements through scripting.

Qosium Scopemon is a measurement controller for automated monitoring. This enables, for instance, continuous monitoring of a critical communications system, allowing immediate reactions if QoS decreases excessively.

A custom controller can be implemented by using the QMCP interface. Kaitotek offers a straightforward QMCP control API for all their customers, enabling direct Qosium measurement control, e.g., from a network management solution.

Results systems

The task of a results system is to gather and store the measurement results. A separate results system is not a mandatory part of a Qosium measurement. Still, it becomes handy when the measurement setup is large, and a lot of measurement results need to be stored continuously. In some cases, an advanced measurement controller also integrates a results system. Results systems may also incorporate centralized visualization and analysis functionalities.

Qosium Storage is a dedicated database server optimized for storing Qosium measurement results. Qosium Storage is especially useful when results data grows large, and accessing the data is wanted to be centralized.

A custom results system can be implemented by using the QMCP interface. Kaitotek offers a straightforward QMCP listener API for all customers, enabling direct Qosium results collection, e.g., for an existing cloud solution.

Support for QoE

Qosium Probe is a QoS and traffic flow measurement solution, but it also includes algorithms for evaluating QoE based on QoS parameters measured. Scientifically proved and user-test-based algorithms for accurate QoE evaluation of video and VoIP are integrated. In addition, a generic QoE algorithm is included, which allows users to specify their own QoS thresholds for obtaining a single QoE value about current quality. No more monitoring of different QoS parameters separately, but the actual quality can be seen through a single value.

Measurement Results

All statistics are collected in both directions, meaning that sent and received traffic can be evaluated separately. The results can be averaged or evaluated accurately per packet.

The most commonly used statistics of Qosium include:

QoS statistics:

  • Packet loss
  • Delay
  • Jitter (delay variation)
  • Connection break statistics

Traffic statistics:

  • Load
  • Volume of data
  • Packet sizes

Flow statistics:

  • Traffic flows traversing measurement node
  • Per flow: load, duration, transmitted data, etc.

QoE (pseudo-subjective analysis)

  • VoIP and video quality with PSQA in the MOS scale
  • Generic QoE algorithm, which allows users to define their own QoS threshold values



In addition, Qosium measures many other parameters and optionally has a full remote packet capturing capability.