百度智能云

All Product Document

          CDN

          Data of Top Class

          Calculation Method

          All TOP data are calculated offline from the access logs, and all kinds of TOP data within 1 hour are counted, with the specific methods shown as follows:

          1. Gather the TOP data of hourly granular according to the log, and take and store TOP 200 data.
          2. Generate TOP data in one-day granularity based on the aggregation of TOP 200 data per hour.
          3. Select 20-50 pieces of data with the largest value to display on the console.

          Note: Because the domain names contained in the TOP 200 are different each hour, resulting in natural differences in the TOP data of granular for a day and the dotting statistical data per minute obtained through aggregation. Therefore, the TOP analysis data are only severed as directional data to help you to better understand the traffic of the domain name, but do not guarantee that the data are consistent with the statistical dotted data value.

          Query Delays

          As TOP data query has a latency of 2-3 hours as it is obtained by calculation of access logs offline.

          Query Granularity and Time Span

          Currently, the query time span supported is nearly 30 days. Only the data of 1-hour granularity can be output.

          Other

          • All TOP data are obtained by offline calculation of the access logs of hourly level as a whole, so the accuracy of TOP is measured in hours.
          • TOP URL and TOP REFERER are both calculated based on the order of access count.
          • TOP HTTPCODE is obtained by ranking and calculating the occurrence of all status codes under the current domain name or user.
          • TOP DOMAIN is obtained by ranking and calculating the traffic generated by all domain names under current user.
          Previous
          Data of Hit Rate Type
          Next
          Monthly Service Report