Thanks, Mark. Would you or @alexcrichton mind sharing more on the DRAM and storage (assuming an SSD of some sort)? Just the SKU name/number would be enough and I can take it from there.
I think there can be some value in reproducing your testing environment as the control and then altering CPU, DRAM, SSD to understand what makes the greatest difference (or not). I could also create another baseline but since the community has already aligned around metrics from perf.rust-lang.org, all the better to try to recreate!
Really like what @nrc did in that blog. Will definitely use it as a reference point.