Hi Influx,

Right now for performance graph we use a format that's the inverse of 
percentiles. e.g. We define "great" performance as taking less than 25ms, 
"good" as <50ms, "ok" as <100ms, "bad" as <1s, and "awful" for everything else.

The advantage of this is that it's easy to see "how well are we doing?" because 
you end up with a colourful stacked graph, and humans are pretty good at 
relative area perception. The result looks like this: http://imgur.com/a/MwdfB

The downside is that the queries needed to create these graphs are pretty 
verbose, and I need to use grafana's ability to render stacked graphs with a 
total of 100%.

SELECT count(value) FROM "ui.switch_thread" WHERE $timeFilter AND value <= 25 
GROUP BY time($interval)
SELECT count(value) FROM "ui.switch_thread" WHERE $timeFilter AND value > 25 
AND value <= 50  GROUP BY time($interval)
SELECT count(value) FROM "ui.switch_thread" WHERE $timeFilter AND value > 50 
AND value <= 100  GROUP BY time($interval)
SELECT count(value) FROM "ui.switch_thread" WHERE $timeFilter AND value > 100 
AND value <= 1000 GROUP BY time($interval)
SELECT count(value) FROM "ui.switch_thread" WHERE $timeFilter AND value > 1000 
GROUP BY time($interval)

What I would really like is a function that lets me calculate these graphs 
directly without using the WHERE clause, so I can combine all 6 queries into 
one continuous query, and maintain these graphs directly in Influxdb:

SELECT PERCENTAGE_BETWEEN(value, 0, 25),
              PERCENTAGE_BETWEEN(value, 25, 50),
              PERCENTAGE_BETWEEN(value, 50, 100),
              PERCENTAGE_BETWEEN(value, 100, 1000),
              PERCENTAGE_BETWEEN(value, 1000, 10000000000000)
FROM "ui.switch_thread" WHERE $timeFilter;

Would anyone else be interested in this function, or is there a way of 
calculating it without using the WHERE clause that already exists?

Thanks,

Conrad

-- 
Remember to include the InfluxDB version number with all issue reports
--- 
You received this message because you are subscribed to the Google Groups 
"InfluxDB" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/influxdb.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/influxdb/4ca0848a-decd-461a-92df-9b1960d4ed9c%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to