So,

 

this is really summary, but just in case it can be of use to someone in deciding something, here's a little tidbit.

 

I've just finished parity checking a new, fresh, 4-drive non-pro WD Red (WD80EFZX) array, running of one of the SAS ports of a supermicro xeon-d board.

 

Duration: 16 hours, 35 minutes, 27 seconds. Average speed: 134.0 MB/sec

 

I also have an array of 5 only Seagate Archives (ST8000AS0002) drives running off the sata ports of an X10SL7-F, with a bit north of 20TB of data on it, depending on the day.

 

Duration: 15 hours, 33 minutes, 27 seconds. Average speed: 142,9 MB/s (there's a bit of variance in the parity-checking history, depending on writes - slowest I've seen it is 81,2 , probably due to heavy non-parity checking activity, fastest has been 144,2)

 

Single parity in both cases.

 

Not sure how replicable all of this is, or how it'd relate to arrays with both shingled and non-shingled drives, or if there's some other variable (likely one between the keyboard and the screen), so take it all with a pinch of salt.

That's a little surprising, would expect the WDs to have similar speed, the server where my 8TB Seagates are now also has other sizes, but when it was 8TBs Seagates only my results were similar to yours, last one was 148MB/s (15H01M14).

That's a little surprising, would expect the WDs to have similar speed, the server where my 8TB Seagates are now also has other sizes, but when it was 8TBs Seagates only my results were similar to yours, last one was 148MB/s (15H01M14).

 

Yeah, I would've expected them to be a touch quicker, but it's close enough to not really be high on my worry list. I'll split up the ports when I get around to it and see what happens the next time it's time for a parity check.