lujiajing1126 commented on issue #9930: URL: https://github.com/apache/skywalking/issues/9930#issuecomment-1596133410
In order to verify TopN Aggregation, we need - scripts to write data - scripts to query TopN in two approaches, i.e. full-scan and preaggregation the infra can be found https://github.com/lujiajing1126/banyandb-billy. The detailed setup can be found there. The metric we are writing is basically, `temperature{sensor_id="<seq_id>"} <temperature>`. It means the measure schema only has a single tag, i.e. `sensor_id` which is a number within a sequence while the temperature (int) is randomly generated. The sequence range is used to control the actual cardinality of the data points. And we have the following benchmark report, | Cardinality | FullScan | PreAggregation(1k counter) | | ----------- | -------- | --------------------------- | | 5K | 2.5 s | 1.0 s | | 10K | 3.6 | 1.0 | | 25K | 10 | 1.0 | | 50K | 19.9 | 1.0 | | 100K | 40 | 1.0 | Lower is better. (Tested on Mac Studio M1 Ultra with 64 GB memory) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
