TY - GEN
T1 - QoE Analysis of the Setup of Different Internet Services for FIFO Server Systems
AU - Hoßfeld, Tobias
AU - Varela, Martín
AU - Heegaard, Poul E.
AU - Skorin-Kapov, Lea
N1 - Publisher Copyright:
© 2018, Springer International Publishing AG.
PY - 2018
Y1 - 2018
N2 - Queueing systems following a first-in-first-out (FIFO) strategy are well understood and various results are known for the response time of the system. However, the question arises how the results look like when taking into account a user-centric point of view. To this end, an M/M/1-FIFO system is investigated for the setup of different Internet services. In this tutorial paper, the impact of the system’s delay on Quality of Experience (QoE) is considered for (1) YouTube video (initial playout delay), (2) authentication in social networks, (3) wireless 3G Internet connection setup. Existing QoE models are used to map the response time in the system, corresponding to the waiting time for users until the service is setup, to Mean Opinion Scores (MOS) as a measure of QoE. The system is then evaluated in terms of overall QoE and QoE fairness for the three services considered, under different load scenarios. The results show how different such systems and response times are perceived by users of different services. Further, the dimensioning of FIFO systems with respect to QoE only requires us to consider the overall QoE.
AB - Queueing systems following a first-in-first-out (FIFO) strategy are well understood and various results are known for the response time of the system. However, the question arises how the results look like when taking into account a user-centric point of view. To this end, an M/M/1-FIFO system is investigated for the setup of different Internet services. In this tutorial paper, the impact of the system’s delay on Quality of Experience (QoE) is considered for (1) YouTube video (initial playout delay), (2) authentication in social networks, (3) wireless 3G Internet connection setup. Existing QoE models are used to map the response time in the system, corresponding to the waiting time for users until the service is setup, to Mean Opinion Scores (MOS) as a measure of QoE. The system is then evaluated in terms of overall QoE and QoE fairness for the three services considered, under different load scenarios. The results show how different such systems and response times are perceived by users of different services. Further, the dimensioning of FIFO systems with respect to QoE only requires us to consider the overall QoE.
UR - http://www.scopus.com/inward/record.url?scp=85042540337&partnerID=8YFLogxK
U2 - 10.1007/978-3-319-74947-1_16
DO - 10.1007/978-3-319-74947-1_16
M3 - Conference contribution
AN - SCOPUS:85042540337
SN - 9783319749464
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 234
EP - 248
BT - Measurement, Modelling and Evaluation of Computing Systems - 19th International GI/ITG Conference, MMB 2018, Proceedings
A2 - German, Reinhard
A2 - Hielscher, Kai-Steffen
A2 - Krieger, Udo R.
PB - Springer Verlag
T2 - 19th International GI/ITG Conference on Measurement, Modelling and Evaluation of Computing Systems, MMB 2018
Y2 - 26 February 2018 through 28 February 2018
ER -