A modest performance improvement is also visible in the average session throughputs, in most of the simulated session arrival rates.
Thus, overall, there will be a smaller percentage of sessions with a low throughput, resulting in a higher average session throughput.
The main KPI depicted is average user session throughput, which is the average throughput of a user's session averaged over all sessions in the simulation.
Figure 8 shows the average and 10th percentile user session throughput as well the average session percentages per layer of the different LC policies and the baseline.
In Fig. 2 b, c, the mean number of active UEs against the session throughput is represented in two cases: when the cell bandwidth is omitted, Fig. 2 b, and when it is included in the definition, Fig. 2 c.
λ a denotes the mean session arrival rate.
The goal is to minimize the values of these metrics, except for session throughput, that should be maximized.
In our previous work [9], we proposed a way of measuring the quality of an admission control mechanism based on the trade-off between the number of servers used and six important QoS metrics: number of overloaded servers, session throughput, number of aborted sessions, number of deferred sessions, number of rejected sessions and average response time for all admitted sessions.
The mean session length of these completed sessions was 23 minutes.
The mean session duration was 45 minutes (SD ± 29.1 min).
Finally, the per-population mean response times are weighted by their corresponding probabilities and session throughputs to obtain a mean response time estimate for the application when it is subjected to the session trace.
