Memory upgrades -- how to benchmark?

[Europa2010AD]

My Mini XL+ came with 64GB of memory, and I just purchased another 2 sticks of 32GB memory (matching ones to the original). What are some benchmarks / parameters I should be looking at, to see any increase in performance before & after?

 

[ChrisRJ]

A synthetic benchmark is of little practical value, at least usually. The ones that matter are your workloads. So I would build list with those that are performance critical and check them.

 

[Davvo]

Output of arc_summary.

 

[Europa2010AD]

Output of arc_summary.

Any specific parameters within arc_summary that I should be looking at to compare before & after I install the additional memory?

Right now, at the base 64GB of memory, my ARC size is at around 82%, with a hit ratio of 99.8%, and "data demand efficiency" at 80% (whatever this is).

 

[Davvo]

Any specific parameters within arc_summary that I should be looking at to compare before & after I install the additional memory?

Right now, at the base 64GB of memory, my ARC size is at around 82%, with a hit ratio of 99.8%, and "data demand efficiency" at 80% (whatever this is).

I would also take a look at the data type composition and the likes.
Still, it's hard to find a significant meaning in those numbers beside a general improvement of the cache size (and thus performance when fetching the data within).

 

[Davvo]

Also take a look at the following thread and the jira link within.

Odd ARC Memory Usage Behavior in 13.0-U4

@morganL suggested that I make a separate thread for this: Here is an illustration of what happens with the zfs arc in u4 for me, is any one else seeing anything like this? (the gap is the reboot for upgrade): You can also see that though there is no memory pressure the arc target drops during...

www.truenas.com