|
|
| ¡¡ Quantized Evaluation for User Aware
Availability Abstract£ºTo quantitatively analyze the reason of Web service failures and performance
degradation£¬ a model of the user perceived availability in cluster based Web servers is
proposed£® Firstly£¬ the system state probability is estimated by Markov birthª²death
process£¬ and then the request loss probability in stable system states is estimated by
queuing theory£¬ lastly£¬ combining both of them£¬ therefore the expression for the
user perceived availability is established to quantitatively evaluate the sensitivity of
user perceived availability to different system parameters£® The numerical results show
that the dominant factors that impact the user perceived availability are the ratio
between the mean detecting time and the mean faultª²free time in low workload and the
ratio between the mean repair time and the mean faultª²free time in high workload£®
Meanwhile£¬ the user perceived availability can be improved by increasing the number of
backª²end nodes£¬ however£¬ when the number of backª²end nodes is increased to a
certain extent£¬ it becomes steady£® |
|
