Assessment of the distance learning server\'s operation strategies and service capacity in advance
Download 399.25 Kb. Pdf ko'rish
|
e3sconf ebwff2023 06016
R
T (s er vi ce ti m e (i n seco nd )) λ - speed of requests entering the system (in second) E3S Web of Conferences 420, 06016 (2023) https://doi.org/10.1051/e3sconf/202342006016 EBWFF 2023 6 ൌ ఒ (6) In this, L q the average number of queries in the queue is determined, and it is found by the formula (7). ൌ ቀ ఘ Ǩ ቁ כ ൬ ఘ ଵି ഐ ൰ כ బ ଵିఘ (7) In formula (6), P 0 is the probability of no requests in the system, which was determined using formula (8). Ͳൌ ଵ ሾଵାσ ൬ ഐೖ ೖǨ ൰ାቀ ഐ Ǩ ቁכሺ భ భష ഐ ሻሿ ೖసభ (8) Fig. 4. Variation of service time in relation to request intensity in the M/G/m/k model It should be mentioned that based on this M/G/m/k model, requests are first sent to the waiting part of the system and then to the service devices when the service devices in the system are free. Therefore, when the service time is estimated based on this model, the system's service time for requests did not increase significantly, even when the number of requests increases by a large margin. Because, in this case, most of the time spent on servicing requests is on standby devices, this time has grown in the same way as shown in Figure 4 above. This affected the average service time and caused an increase in the service time. At the same time, the number of waiting areas also became significant in the service process, and whether they were limited or infinite had an effect on the service time. Based on [20], it can be said that the service time is also finite in the service with finite waiting areas, and this method is a conditional loss service process. Due to this, the availability of service devices and waiting areas causes a loss of incoming requests at this time. However, having infinite standby devices also negatively impacts service time and causes requests to be queued. This, in turn, produces an endless increase in service time as a result of a significant increase in the number of requests [20]. User Request Web server Server Respond for request Queue Internet Fig. 5. Service requests based on the M/G/m/k model 0 0.5 1 1.5 2 2.5 3 3.5 4 10 41 69 100 130 161 191 Download 399.25 Kb. Do'stlaringiz bilan baham: |
Ma'lumotlar bazasi mualliflik huquqi bilan himoyalangan ©fayllar.org 2024
ma'muriyatiga murojaat qiling
ma'muriyatiga murojaat qiling