See this thread: understanding the graphs
From what I understand, those graphs show a healthy alignment ratio for all drives, and logs will roll and rotate so I'm not terribly surprised if that data isn't entirely aligned at a block level either.
Do you have other heavy IO VMs on the same VxRail host(s)? Also what does the network utilization on the backend look like?
Yes, I came across that thread. Thank you.
There are some DB VMs that have heavy workloads on this vSAN Cluster, that have similar IO behavior. According to Network Insight, there's nothing suspicious when it comes to utilization and/or throughput. I don't understand why wouldn't the log rotation just align. Is there way to find out the physical sector size of an SSD in esxcli? I tried with "esxcfg-info" but it doesn't give that info or I don't see it.
Just in case you want to look at all the graphs of DB, Log and TempDB, here it is,