I too find the lack of time period information confusing. However by clicking in small increments along the graph it seems that:
- 5 minutes: Data per second.
- 3 Hours: Data totalled in five minute buckets. Divide by 300 to get average data per second for each five-minute period.
- 24 Hours: Data totalled in hour buckets. Divide by 3600 to get average data per second for each hour.
- Week: Data totalled in six hour buckets. Divide by 21,600 to get average data per second for each six hour period.
- Month: Data totalled in day-long buckets. Divide by 86,400 to get average data per second for each day.
The issue isnât helped by the bucket totals (which should be solid rectangular bars) have been smoothed to make a curvy area graph. Only the centres of the buckets have the correct Y-axis height, the curved parts donât provide any additional data.
I would suggest the graphs could be significantly improved by:
- Displaying the average data rate per second in the top left corner, whatever the time scale. That is, divide the current number by the number of seconds per bucket.
- Either display the bucket size (e.g. â12KB/s averaged over hoursâ) or display the data as a proper bar graph instead of the smoothed bar graph. Or perhaps display the smoothed line on top of the actual bars?
- Display the bucket time period in the status information under the graph when the graph is clicked. (e.g. â22GB this dayâ, â128KB this five minutesâ, etc.).