Post Reply 
 
Thread Rating:
  • 1 Votes - 5 Average
  • 1
  • 2
  • 3
  • 4
  • 5
read/write performance disparity
05-22-2013, 11:33 AM
Post: #11
RE: read/write performance disparity
Well, it's fixed. For now. Unfortunately I had to format the Drobo and restore all data from backups. Now I'm pulling 300+MB/s read and 220+MB/s writes. Performance in Finder never dips below 100MB/s for common tasks, even writing large batches of small files.

Long story short, this 5D was configured new with 2 3TB drives. After 2TB of data was copied to Drobo, 3 smaller drives (~700GB each) were freed up and I added them to the 5D. It stands to reason that when Drobo took that data from 2 large drives and spread it over 5 drives of mixed size it did so stupidly, resulting in erratic performance. Of course, maybe I'm the stupid one for assuming it would "just work."

Either way, I'm glad the full potential of the device is finally being realized.
Find all posts by this user
Quote this message in a reply
05-22-2013, 01:54 PM
Post: #12
RE: read/write performance disparity
See how it performs in a few days. Like Docchris mentioned earlier, when initially set up, Drobo may not be using all available drives.

If the smaller drives are causing your performance issues, they are likely to return.

At least you know your chassis itself isn't slow though!

--Brandon | WHS2011+Drive Bender/2x Drobo v2/Drobo S G1/ Drobo S G2/Transporter
Drobo provides fault-tolerance, it's NOT a substitute for regular backups.
Drobo Best Practices - Official and Community-sourced.
Find all posts by this user
Quote this message in a reply
05-23-2013, 08:36 AM
Post: #13
RE: read/write performance disparity
hmm, now that you mention 4gb zip files...
i remember times when a highly compressed file took longer to create/uncompress, than a file with low compression.
the cpu processing was always slower with higher compression, but you ended up with the file taking up less space.

maybe the level of your compression (or compression overhead) had something to do with it too?

(btw i have XP home SP2, a Drobo v1 with 2x 1TB/2x 1.5TB WD greens, & a bkp Drobo v2 with the same + a DroboShare: unused)
& a DroboS v2 with 3xWD15EADS &2x1TB in DDR mode on win7, & a drobo5D (all usb)
  • btw i did a sustained (write) operation for about 6 hours, and got 13.2MB / sec ...objection? "sustained" :)
    (16.7MB/s on a v2 & 47-96MB/s drobo-s)
Find all posts by this user
Quote this message in a reply
05-23-2013, 09:41 AM
Post: #14
RE: read/write performance disparity
I'm assuming mestanczak is timing straight file copy/transfer, rather than timing the opening/interpreting of files from the 5D, but it's a good clarification point.

From a storage point of view, the read/write speed of compressed vs non-compressed data should be the same, unless the host or device itself is applying compression.

--Brandon | WHS2011+Drive Bender/2x Drobo v2/Drobo S G1/ Drobo S G2/Transporter
Drobo provides fault-tolerance, it's NOT a substitute for regular backups.
Drobo Best Practices - Official and Community-sourced.
Find all posts by this user
Quote this message in a reply
05-23-2013, 11:17 AM
Post: #15
RE: read/write performance disparity
ah yes it could be just a straight one i wasnt sure Smile

in nfts, you can have files / folders in "blue" denoting nfts compression.. i've used some of that on my drobos, but it seems to work fine even when playing back large half gig video files which are blue. i havent tried that with hundreds of files though but im assuming that would add some overhead.

(btw i have XP home SP2, a Drobo v1 with 2x 1TB/2x 1.5TB WD greens, & a bkp Drobo v2 with the same + a DroboShare: unused)
& a DroboS v2 with 3xWD15EADS &2x1TB in DDR mode on win7, & a drobo5D (all usb)
  • btw i did a sustained (write) operation for about 6 hours, and got 13.2MB / sec ...objection? "sustained" :)
    (16.7MB/s on a v2 & 47-96MB/s drobo-s)
Find all posts by this user
Quote this message in a reply
05-24-2013, 09:50 AM
Post: #16
RE: read/write performance disparity
Native compression in Windows is fairly "loose" to both not adverse affect performance and I don't think it's changed much since NT, which ran on CPUs that are two decades old and probably slower than a desktop calculator (definitely slower than a current TI calculator).

--Brandon | WHS2011+Drive Bender/2x Drobo v2/Drobo S G1/ Drobo S G2/Transporter
Drobo provides fault-tolerance, it's NOT a substitute for regular backups.
Drobo Best Practices - Official and Community-sourced.
Find all posts by this user
Quote this message in a reply
05-31-2013, 10:08 AM
Post: #17
RE: read/write performance disparity
good old texas instruments Smile

(btw i have XP home SP2, a Drobo v1 with 2x 1TB/2x 1.5TB WD greens, & a bkp Drobo v2 with the same + a DroboShare: unused)
& a DroboS v2 with 3xWD15EADS &2x1TB in DDR mode on win7, & a drobo5D (all usb)
  • btw i did a sustained (write) operation for about 6 hours, and got 13.2MB / sec ...objection? "sustained" :)
    (16.7MB/s on a v2 & 47-96MB/s drobo-s)
Find all posts by this user
Quote this message in a reply
06-01-2013, 03:59 PM
Post: #18
RE: read/write performance disparity
TI-81 got me through HS and College Calc...

i7 Mac Mini Server, 10.10
Drobo (Gen2), USB, 4x WD 4TB RED

Drobo (Gen3), USB3, 4X WD 4TB RED

Drobo 5D, TB, 5x WD 4TB RED, 120GB Intel 525 mSATA SSD
Find all posts by this user
Quote this message in a reply
06-06-2013, 11:38 AM
Post: #19
RE: read/write performance disparity
maybe TI will one day design a Federation PAD Smile

(btw i have XP home SP2, a Drobo v1 with 2x 1TB/2x 1.5TB WD greens, & a bkp Drobo v2 with the same + a DroboShare: unused)
& a DroboS v2 with 3xWD15EADS &2x1TB in DDR mode on win7, & a drobo5D (all usb)
  • btw i did a sustained (write) operation for about 6 hours, and got 13.2MB / sec ...objection? "sustained" :)
    (16.7MB/s on a v2 & 47-96MB/s drobo-s)
Find all posts by this user
Quote this message in a reply
02-15-2016, 02:21 AM
Post: #20
RE: read/write performance disparity
Sorry to revive an old thread, but I'm experiencing the exact same issue as the OP. My read speeds according to Blackmagic are less than half my write, which is very unusual - 70MB/s Read vs 173MB/s Write.

I only started to notice the issue when I moved my iTunes library onto my Drobo and it would take forever to navigate throughout the library. However, to double check, I used a different app called AJA System Test which seemed to be giving me different results - 210MB/s Read vs 190MB/s Write.

I did some digging and apparently Blackmagic uses incompressible data to test where as AJA uses compressible data, and apparently that's what causes the different results. Having read the fix posted by mestanczak, I'm now wondering if I should format the Drobo just to be safe or is my unit functioning perfectly normal as it should?

It pains me to figure out where I can dump 6.5TB temporarily though..
Find all posts by this user
Quote this message in a reply
Post Reply 


Forum Jump: