Protocol-agnostic method for monitoring interactivity time in remote desktop services

Abstract

The growing trend of desktop virtualisation has facilitated the reduction of management costs associated with traditional systems and access to services from devices with different capabilities. However, desktop virtualisation requires controlling the interactivity provided by an infrastructure and the quality of experience perceived by users. This paper proposes a methodology for the quantification of interactivity based on the measurement of the time elapsed between user interactions and the associated responses. Measurement error is controlled using a novel mechanism for the detection of screen changes, which can lead to erroneous measurements. Finally, a campus virtual desktop infrastructure and the Amazon WorkSpaces solution are analysed using this proposed methodology. The results demonstrate the importance of the location of virtualisation infrastructure and the types of protocols used by remote desktop services.

Interactivity Anomaly Detection in Remote Work Scenarios Using LSTM

“…The literature currently presents two primary methodologies for assessing QoE of remote desktops through the quantification of interactivity time: Slow-motion benchmarking [17] [18] and thin client latency analysis (TeCLA) [14]. Both approaches aim to measure the interactivity time of remote desktops.…”

“…• Widely adopted and wellestablished • Easy to implement • Independent of the protocol or application used Thin Client Latency Analysis (TeCLA) [14] • Observing changes on the screen, as well as keyboard and mouse interactions, to identify when screen updates are received by the thin client.…”

“…Among the metrics that characterize QoE, the interactivity time, the number of frames per second, and the image resolution stand out. Interactivity time [14] is the time elapsed between a user interaction via keyboard or mouse and the on-screen update with the result of the user interaction. This metric is also referred to as responsiveness [15].…”

“…The literature suggests that further work is needed on remote work communications [16]. In this work, we build upon the extraction of interactivity time from a previous study [14], using the system in a national parcel delivery company with a network of remote desktops across different offices. We collected data from real users to design a scheme for the automatic detection of anomalies in interactivity time.…”

Survey on Quality of Experience Evaluation for Cloud-Based Interactive Applications

“…One of the main network metrics related to the QoE in CIAs is the time elapsed from when the user interacts with the application until receiving a graphical response. This metric is called the interactivity time [9] or responsiveness [10]. Increased interactivity time can reduce the usability of a CIA.…”