Add Starrocks

[lio-p]

Add Starrocks results to JSONBench.

The results are partial, only 1, 10 and 100 millions rows are present. The 1 billion rows could not be run with this setup, the data load would fail.

[murphyatwork]

hey @lio-p, may i ask what error did you encounter when ingesting data ?

i'm trying to run this benchmark, but didn't see any error when ingesting 100m. maybe some surprises in 1000m test ?

[lio-p]

As far as I remember, everything was fine up to 100 Million documents. But once the 1000M dataset was ingested, the Starrocks backend service will start rebooting on random queries, which made the benchmark impossible to run.

I didn't keep the error message unfortunately.

[murphyatwork]

As far as I remember, everything was fine up to 100 Million documents. But once the 1000M dataset was ingested, the Starrocks backend service will start rebooting on random queries, which made the benchmark impossible to run.

I didn't keep the error message unfortunately.

I see. Thanks for reminding us. Our team will fix it to reach a reasonable number.