I have a time-series data which is ingested daily. I am using TSVB Time Series graph to do a filter ratio calculation to get the percentage of good records (Basically filter for good records divided by Total).
Currently the calculation is done per 1 day (the defined interval) and I see the success ratio on a daily basis on the graph for that day. What I need though, is to be able to calculate the success ratio, everyday over the entire data set since the beginning to present (not just for the past day).
For instance, let's say we started ingesting on Jan 1st, 2020. On Januray 5th, 2020 I need to see the success percentage from Jan 1st to Jan 5th on the graph. On Jan 6th, I need to see the success percentage from Jan 1st to Jan 6th on the graph. On present day I need to see the result from Jan 1st to today (and so on).
Could anyone suggest a way to do this using either Timelion or TSVB? I prefer TSVB since I am more familiar with it.
Thanks for the response... I actually tried that but it doesn't solve my issue. The actual calculation is always performed as per the interval defined inside the TSVB graph (in my case for the past 24h). Changing the date in the time picker only narrows down the visibility of the graph... it does not impact the calculation interval.
I think what I would need is to be able to define the "From Jan 1st TO now" interval inside the visualization itself but I don't think that is possible?
You mean the whole bucket that you calculate the success has to change dynamically as well? That isn't possible, sadly. i mean, you could use some kind of cumulative aggregation that will give you the same result. Something like "cumulative sum of success", "cumulative sum of failure" and then do a ratio on them and you'll get the success rate for the whole graph.
Apache, Apache Lucene, Apache Hadoop, Hadoop, HDFS and the yellow elephant
logo are trademarks of the
Apache Software Foundation
in the United States and/or other countries.