Parity Checks Bus limited?

[bnevets27]

Thanks again Johnnie, your help here is very much appreciated, as was everyone else's. 

Looks like I should be getting a pretty decent bump with the new hardware. I shouldn't have any bottlenecks other then a bit of CPU, which I'm ok with. 

 

Hopefully if someone is having a similar problem this will help them figure it out. Will be really nice if/when unraid can do parity calculations in multithread. 

 

 

[UhClem]

A week and a half ago ...

On 12/11/2018 at 6:57 PM, UhClem said:

... you can use dskt2 (just specify the full list of drive-letters). In either case, if the ~1GB/s ceiling isn't replicated, a remaining candidate is the CPU, [if unRAID's 2-parity implementation is "uninspired" and single-threaded] -- check with mpstat (I think).

 

To which, you replied:

 

Quote

Ran the following command
 

dskt2 s m r k h g f e l p n o d c b

This is the result

sds = 164.02 MiB/sec

sdm = 179.66 MiB/sec

sdr = 151.12 MiB/sec

sdk = 148.25 MiB/sec

sdh = 115.86 MiB/sec

sdg = 133.69 MiB/sec

sdf = 131.94 MiB/sec

sde = 140.57 MiB/sec

sdl = 109.92 MiB/sec

sdp = 164.80 MiB/sec

sdn = 100.93 MiB/sec

sdo = 148.13 MiB/sec

sdd = 143.46 MiB/sec

sdc = 126.61 MiB/sec

sdb = 175.05 MiB/sec

 

 

 

That 10-second test demonstrates that your system supports a concurrent bandwidth of at least 2100 MB/s; far exceeding your observed 1000-1250 MB/s bottleneck.  Therefore, the "~1GB/s ceiling" is not replicated, and, as I stated, the remaining candidate is your CPU (based on 2-parity, single-thread, and an "uninspired" implementation).

 

Regardless, you stated:

Quote

I'm running dual L5420's so while not modern, more than enough for parity sync.

Lesson: When you are digging deep and trying to solve a perplexing puzzle (such as you were--in this thread), it is very important to question ALL of your own assumptions, for that is where blind spots lurk.