Skip to content

Architecture overview

Main components

The Kadiska solution is a SaaS-based DEM (Digital Experience Monitoring) platform that is composed of three main components:

  • The Platform
  • The Watcher
  • The Tracer

Screenshot

The Platform

The Kadiska Platform consists of a series of components that enable you to fully configure and monitor your environment from a unique user interface. It is accessible at https://app.kadiska.com. See here how to get access to it.

The Watcher

The Kadiska Watcher monitors real users activities by collecting metrics from their web browsers. This is achieved through either the use of a specialized RUM JavaScript, called App-Watcher or a browser's extension, called User-Watcher. They instruct the browser to send performance traces to the Kadiska Platform.

The Tracer

The Kadiska Tracer consists of a series of synthetic tests that are launched from Kadiska Stations. You can make use of the provided global network of Managed Stations or deploy your own Enterprise Stations. Today, Kadiska offers the Net-Tracer as a first type of synthetic test.

Deployment model

The way a typical Kadiska total monitoring solution works is schematized hereunder:

Screenshot

First, configure a Kadiska Watcher that corresponds to the digital service to monitor. In an internal web service scenario, you will probably use the App-Watcher to collect performance metrics from the users' browsers. In a SaaS monitoring scenario though, you would use the User-Watcher instead. Optionally, you can configure multiple Tracers to actively test this digital service, by using Managed Stations and/or your own Enterprise Stations.

Once this is done, when a user connects to this digital service, the browser automatically downloads the Kadiska App-Watcher from the Kadiska CDN (Content Delivery Network) infrastructure or activate the User-Watcher according to the configured Watcher method. It enables the browser to collect performance and usage metrics that are then sent to the Kadiska Platform. These metrics are not only related to connections to the digital service itself, but also to any call to third-party services providers, like DNS and CDN. At this stage, the Kadiska embedded Machine Learning analyses the data and triggers the alerting system in case of any significant abnormal behavior or deviation from the baselines. When degradations occur, additional Tracers can be automatically activated to perform additional tests to all parties involved in delivering the digital service. These tests are performed by Stations that are the closest to the users that are experiencing the degradations.


© 2022 Kadiska | Digital Experience Monitoring DEM. All rights reserved