Printable View
http://www.xtremesystems.org/forums/...=1#post4932126Quote:
Originally Posted by B.A.T
Nah, the drive could still perform some writes, but because of the condition of the NAND, the controller kept crashing.
Thus it was not returning the correct errors for read-only mode. There is a specific error to return to say the device is readonly, and crashing the controller isn't the correct method :)
I'm working on finding the correct reference material again, but drives becoming read only is for the most part a decision the controller makes. Most drives only become read only once all/most of the reserve blocks are gone after the controller write protects the drive. We never get to see this behavior because the drives are all dying from something else.
I found one instance again --
From a Micron Product PDF for the P400e:
It's not something that happens organically, it's something the controller decides to do in the event there are no reserve blocks left. Also, I believe SF does this as well. So while Read Only is possible, it's practically never going to happen. Unless maybe the X25 V starts dropping several blocks every day for the next couple of years.Quote:
Total Spare is the block count used to substitute (replace) bad blocks before the drive
goes into write protect mode. It is a fixed value.
PercentageLifetimeUsed is calculated as:
PercentageLifetimeUsed = max(PercentageEraseCountUsed,
PercentageSpareBlockUsed)
Attachment 127471
Won't be long now. LBA writes are 13TiB ahead or so, but either way 2PiB isn't far off.
That is one quick writer :shocked:
When the 830 reaches 3PiB, that is when things will probably start to get a bit perillous!
Man that Sammy is smokin'. Just makes me wonder just how much the 512MB 830 could rack up. Can't imagine someone sacrificing that big of a drive tho...I couldn't.
well done :up:Quote:
Originally Posted by B.A.T
the retention test is a good idea, don't forget to update to fw 0309 (or 000f)
to prevent the "5K hours bug"
bye
luke
Intel 330 120GB
172.83TB Host writes
Reallocated sectors : 05 0
Available Reserved Space : E8 100
MWI 81
[F1] Total LBAs Written 5663293
[F2] Total LBAs Read 30852 // not much use in this one
[F9] Total NAND Writes 124755GB
POH 452
MD5 OK
122.95MiB/s on avg (~11 hours)
The 330 has been on and off as I needed that port for updating fw on the Vertex 4's.
It was idling for a few hours while the X25-V joined as a member of the PiB club :)
--
Kingston SSDNow 40GB (X25-V)
1026.18TB Host writes
Reallocated sectors : 05 46
Available Reserved Space : E8 99
POH 9180
MD5 OK
37.31MiB/s on avg (~11 hours)
--
Attachment 127474
Attachment 127475
I'm planning on having a retention test next weekend (Friday-Sunday) on both drives, it will be the first for the 330, the V has had quite a few tests but it's been a while since last time. (probably at 800GiB)Quote:
Originally Posted by B.A.T
An SSD getting into a true Read-Only state is what we have been "told" or told our-selves would happen, it does not look to happen in the real-world.
Even if it did, data-retention would make unplugging the drive a high risk sport.
On the other hand, if there is a chance that someone implemented a Read-Only state I'd say the odds are good on Intel and Micron.
So, it's a tough one, I'm set on performing a test next weekend, it may end in "tears" though :)
(if it makes it through the test there will be no more retention testing on the V)
If there were a true readonly mode ... the One_Hertz's Intel 320 drive probably should have triggered it ... It used all its spare space ... then some!Quote:
Originally Posted by Anvil
Maybe the Sandforce based drives (including intel 330/520) might go readonly when reserved space is exhausted ... but these drives don't seem to be able to properly identify blocks that are too worn and should be removed from service ... so important metadata ends up on them and the data disappears during downtime ... or the drive just up and panic locks! (The sandforce drives are quite well designed, but lack of proper testing tended to negate the design quality, maybe it has improved with LSI at the helm!)
More investigation needed *evil grin*
One_Hertz drive was giving warnings (they all did in some way) so under normal circumstances one would simply replace the drive, it should have triggered read-only mode long before it finally died. (so, getting a drive into read-only mode looks to be impossible)
I've got plenty of drives... :)Quote:
Originally Posted by canthearu
Intel 520 60GB - Day 105
Drive hours: 2,466
ASU GiB written: 783,153.21 GiB (764.80 TiB)
Avg MB/s: 93.62 MB/s
MD5: OK
Host GB written (F1): 788,814.75 GiB (770.33 TiB, 25242072 raw)
NAND writes (F9): 558,420 GiB (545.33 TiB)
Reallocated sectors (05): 0
Failure count (AB, AC): 0 program, 0 erase
Raw Error Rate (8B): 102 normalised
Avaliable Reserved Space (AA): 100 normalised
Media Wearout Indicator (E9): 1
Sandisk Extreme G25 - 120gb - Day 5
Drive hours: 118
ASU GiB written: 49,603.28 GiB (48.44 TiB)
Avg MB/s: 120.47 MB/s
MD5: OK
Host GB written (F1): 50,075 GiB (48.90 TiB)
NAND writes (E9): 37,379 GiB (36.50 TiB)
Retired Block Count (05): 4
Failure count (AB, AC): 1 program, 1 erase
Raw Error Rate (01): 104 normalized
Media Wearout Indicator (E7): 93 normalized
Wear Range Delta (B1): 4 raw
Reported Uncorrectable Errors (BB): 0 raw
No further re-allocations on the G25.
I'll update the charts later tonight.
Interesting development on the SanDisk, re-allocations sort of resemble the X25-V. (early days though)
Intel 320
Some of you may vaguely remember way back when (last summer??) I posted Intel pdf links for the Intel 320.
I had found in one of the Intel pdf's the Intel 320 had no read only point as some of the SSD's reportedly do.
It would write until it died...thus it did. Alas, poor Intel 320 we knew thee...
Without the controller stepping in and declaring the drive read-only, you'd never see RO otherwise.... it just doesn't work like that. One manufacturer even measures life by % of the way to RO. But we'll never see it here, it could be more common if we were doing full-span random testing. Either way, for a drive to become RO it has to have no reserve space, and that's just not going to happen for us. Most will simply wake up dead one day with very few reallocations. Others, like the two Turbos, will have one die too damaged to continue (you could see it with the Turbos).
I only know two instances where drives had become RO, one was a defective drive on a manf. support forum which had 200 hours and no reserve space left, while another instance was a SF PCIe drive (it was detailed here in a thread IIRC). With millions of SSDs deployed, you'd think there would be more anecdotal evidence of drives hitting RO conditions.
--------------------------------------------
Plextor M3P 128GB Day 13
336,892.18 GiB
328.00 TiB
272.29 MB/s Average
348 Hours
Rellocated Event Count 0
--------------------------------------------
Samsung 830 256GB Day 83
2,018,309.13 GiB
1,971.00 TiB
294.56 MB/s Average
1994 Hours
8763 Wear Leveling Count
MWI 1
7/0 Erase/Program Fail
Used Reserved Block Count: 14/28672 sectors
---------------------------------------
Thank you :)Quote:
Originally Posted by 2f_tfe
I'm going to wait it out until I'm closer to 5K hours. With so much P/E used I might not see 5K hours.
Ok. I'll do something similar. I'll start the retention test today with both the G1 and m4. It'll be 12, 24 and 48h and it'll be the last for the m4 while the G1 will get more with 200 TiB intervals for now.Quote:
Originally Posted by Anvil
Todays update:
m4
1037.9935 TiB
4355 hours
Avg speed 74.55 MiB/s.
AD gone from 21 to 16
P/E 17882.
CE 26
MD5 OK.
Reallocated sectors : 00
http://www.diskusjon.no/index.php?ap...tach_id=493981
Intel X25-M G1 80GB
705.33 TiB
23665 hours
Reallocated sectors : 152 to 158
Available Reserved space: 75
MWI= 44
MD5 =OK
38.90 MiB/s on avg
http://www.diskusjon.no/index.php?ap...tach_id=493980
Intel X25-E 64GB
246.80 TiB
753-30=723 hours
Reallocated sectors : 0
Available Reserved space: 100
MWI= 98
MD5 =OK
92.13 MiB/s on avg
http://www.diskusjon.no/index.php?ap...tach_id=493982
Mtron Pro 7025 32GB
ASU write 27.0836 TiB
Host write 49,5474 TiB
Total host write 97,0830 TiB
Day 8
MWI= 98 to 97
MD5 =OK
47.08 MiB/s on avg
http://www.diskusjon.no/index.php?ap...tach_id=493983
The Mtron dropped another point of MWI. I looked in the log and it's 30 TiB of host write for 1% drop. That means MWI=1% will take 3 PiB of host write and 1,5 TPiB of ASU write. The WA is almost 2:1 at the moment.
Intel 330 120GB
182.58TB Host writes
Reallocated sectors : 05 0
Available Reserved Space : E8 100
MWI 80
[F1] Total LBAs Written 5982882
[F2] Total LBAs Read 30852 // not much use in this one
[F9] Total NAND Writes 131796GB
POH 475
MD5 OK
122.28MiB/s on avg (~34 hours)
--
Kingston SSDNow 40GB (X25-V)
1028.83TB Host writes (33712720*32)
Reallocated sectors : 05 46
Available Reserved Space : E8 99
POH 9203
MD5 OK
34.32MiB/s on avg (~34 hours)
--
Attachment 127491
--------------------------------------------
Plextor M3P 128GB Day 14
357,158.80 GiB
348.78 TiB
271.63 MB/s Average
369 Hours
Rellocated Event Count 0
--------------------------------------------
Samsung 830 256GB Day 84
2,039,545.02 GiB
1,991.74 TiB
291.56 MB/s Average
2016 Hours
8855 Wear Leveling Count
MWI 1
7/0 Erase/Program Fail
Used Reserved Block Count: 14/28672 sectors
---------------------------------------
1.96PiB LBA writes on the 830.
Maybe the 830 will make it to 3TB...
I'm losing 850GiB a day, or 5.9TiB a week, with the 830 in the endurance rig. It's just 10MB/s avg less a second (vs. my main system). Not a huge loss, but an irritating one. I've tried everything I could think of to get the H67's speed up, but to no avail. The 830 is just being held back by the platform. The M3P is only a couple MB/s less on average though, so it's not a big concern.
You're using the Intel controller on the H67, correct? What is your main system?
It's a Z68. Both are on the same RST drivers. The Z68 I'm using just has better storage performance, period. I can't necessarily explain it all, but it's just faster than the H67 and P67s I've been using.Quote:
Originally Posted by Kain665
I've heard more than a couple of folks say that the Z68 is edging out the new Z77s as well...strange...not that I'm complaining lolQuote:
Originally Posted by Christopher