Thank you for the concern and interpretation. I get the reason. So I think meta data write in array update interface would take less time than the vos_perf test before due to less data would be written in SCM. Is there something wrong with my understanding? Looking forward to the benchmark result.
