A percentile is the value of a variable below which a certain percent of observations fall. For example, the 95th percentile is the value (or score) below which 95 percent of the observations are found.
95th percentile monitoring relates to bandwidth. This statistic is useful in measuring data throughput because it more accurately reflects the required capacity of the monitored link for applications that are bandwidth sensitive. The 95th percentile says that 95 percent of the time, the bandwidth usage is below this amount. The remaining 5 percent of the time, the bandwidth usage is above that amount. When using 95th percentile to perform capacity planning, we recommend setting the poll interval to at least 1-minute intervals for the monitored devices.
The 95th percentile is calculated for rollups and for reporting purposes.
A rollup is the process during which metric values are aggregated. In an hourly rollup, the 1 minute, 5 minute, 15 minute, 30 minute, and 60 minute polled values for metrics are aggregated every hour. In a daily rollup, hourly values for metrics are aggregated once a day. In a weekly rollup, daily values for metrics are aggregated once a week.
|
Copyright © 2015 CA Technologies.
All rights reserved.
|
|