parallel load data

[a615549958]

Hi, I use influxdb-comparisons which tool you developed. In load_data.sh file ,I set workers>1, where is "broken pipe" error running. I find "late write" in log. Ｂecause I loading data's timestamp is smaller than database data? If use parallel loading ,this error will still exist. Do you have good idea for parallel loading.

[Lazin]

This can happen if you haven’t cleared the database between the runs of the ingestion tool from the influxdb-comparisons. It'll try to write the same data twice. And, I believe that the 'influxdb-comparisons' doesn't work in parallel. I have this set of scripts that can be used to write data in parallel, if you want to benchmark the top ingestion performance - https://github.com/Lazin/roundtrip_test. I think it's easier to use than the tool you mentioned.

BTW, I'm planning to add Akumuli support to the https://github.com/timescale/tsbs test suite. It will be easier to use and it will work in parallel.

[a615549958]

Thanks, I'll have a try.