SSD Write Endurance 25nm Vs 34nm

[Anvil]

The Force 3 lasted 10 minutes or so and now it's not detecting

Will give it an hour or so before I retry, it was at 1015.nTiB when I restarted, unfortunately the logger had been off since last restart.
(not sure why but I expect it's like that old saying, when it rains, it pours )

[devsk]

The Force 3 lasted 10 minutes or so and now it's not detecting

Will give it an hour or so before I retry, it was at 1015.nGiB when I restarted, unfortunately the logger had been off since last restart.
(not sure why but I expect it's like that old saying, when it rains, it pours )

Looks like its on its way out.

[Anvil]

It looks that way, will retry later tonight.
It had been sitting "idle" (idle as in dead/not counting hours) for 7 hours when I discovered the issue, it was discovered without any issue the first time. I've tried on several systems and now there is nothing.

[Ao1]

How did you do the check sum of filled data? Did you fill the disk with a large file right before MWI exhausted? I found that doing an SHA-1 or MD5 hash value of the whole physical disk doesn't work, as each time I mount the disk, the operating system would always write something into the disk...

I created three folders and filled them with mp’s, docs and apps. I check summed each folder using “filecheckmd5” Hopefully it works.

[Christopher]

Surely the F3 is not really that bad yet. What is attribute 1's worst value?

The F3 has very nearly like 2x the raw writes and 2x the host writes, but it 'disconnected' for the last time while writing.

The 830 is near where the Mushkin died in terms of NAND writes too.

[Anvil]

I checked a minute ago and there are no signs of life on the F3.

I'll try a few things and if they don't work then it is over at 1015TiB.
The smartlogger was unfortunately not enabled and as I was trusting the utility I have only myself to blame for not checking that it was enabled.
(I did not save any other screenshots)
Last log entry was 2 days ago but it doesn't really matter, next time I'll make sure to save the daily screenshot and another one when there are issues

The way it was disconnecting resembled the fw issue and so I mistakenly took it for a SF bug but it clearly wasn't, so it looks like one must take such disconnects more seriously as of now.

It was 24 hours from going past 1PiB, I find that more than annoying
Well, the goal is still out there...

[Anvil]

Kingston SSDNow 40GB (X25-V)

664.57TB Host writes
Reallocated sectors : 05 21
Available Reserved Space : E8 99
POH 5992
MD5 OK

33.09MiB/s on avg (~4 hours)

--

Corsair Force 3 120GB

01 120/50 (Raw read error rate)
05 2 (Retired Block count)
B1 42 (Wear range delta)
B6 1 (Erase Fail Count)
E6 100 (Life curve status)
E7 10 (SSD Life left)
E9 771069 (Raw writes) ->753TiB
F1 1026074 (Host writes) ->1002TiB (~22TiB left until 1PiB is reached)

MD5 OK

---.--MiB/s on avg (~-- hours)

power on hours : 2917

--

From what I recall

B1 Wear Range Delta 39
Host Writes 1015.n TiB

[Christopher]

Wow. 01/RRER was at 120? That's curious.

I don't think it was an endurance-related failure in the truest sense. And now I question whether the Mushkin really died from bit errors, or whether it died like the F3 as well - it just up and disconnected on the Morning of day 51 while writing.

I think we need another SF in the test....

[Anvil]

I'll have a look at the data, I don't think there were RRERs that were close to the threshold.

I was hoping for more as well but 1PiB is a lot of data and there virtually are no breaks.

edit:

I had to log 120 as I was reporting while the drive was virtually just re-connected, the entry above is 2 days old.

I've got another F3 but was thinking of running at a different level of compression.
(it's still in the original box, sealed)

I'll leave the F3 until tomorrow and see if miracles can still happen

[bluestang]

Oh man, that's to bad Anvil. Hope it comes back for ya.

Now I'm even more tempted to put in the other 64GB M225 to beat that mark!

[Christopher]

After 31.5 hours of endurance testing, and another 6 hours before that, the MOBI has finally hit MWI 99.

Looks like the one SMART attribute is erase count/MWI at 100/99/14,459,091. Maybe this won't take as long as I had feared....

[Anvil]

Well, if it had lasted past 1PiB there is always 2PiB

edit:

Christopher

Next time you visit the rig, could you change to 46% on the MOBI

[Christopher]

Well, if it had lasted past 1PiB there is always 2PiB

edit:

Christopher

Next time you visit the rig, could you change to 46% on the MOBI

Done. I put all drives on 46 percent by default [but I had neglected to do so on the MOBI -- it's actually on another system, not the Endurance Rig].

Do you think it should make much of a difference? If anything, it seems ever so slightly faster -- but that should change over the next few hours, settling around 54.40MBs.

[Anvil]

In theory it should not matter at all.

You can put 8% or 25%, the reason 46% was suggested was that the SF drives were using it.

[Christopher]

MOBI 3025 16GB SLC Day 1


TiB 6.8665TiB
GiB 7031.27
54.99MBs Avg
Erase Count:
99/100/15019939




Samsung 830 64GB Update, Day 53
FW:CXM02B1Q

TiB 354.1172
GiB 362616.05
133.84 MB/s Avg
PE Cycles:
16112

Reallocated Sectors
45056
22 Blocks, holding steady

1258 hours




OCZ Vertex Turbo 64 Update Day 9

TiB 53.6819
GiB 54970.28
82.87 MBs Avg

Avg Erase Count
985
MWI 81 down from 83


1 Program Fail
22 Erase Fail (up from 21)
0 Read Fail
200 Hours





UPDATE---->
Whooops! I killed the MOBI (on accident)!
I am so embarassed right now.

I did something really dumb, and I deserve to be shot for my transgression. The MOBI is dead, and it's my all my fault.

I'll get another one.

[Anvil]

12 hours later and the Force 3 is not doing well

Looks like it died during a TRIM. (it did, the screenshot tells the story)

Will be updating the stats later tonight.

@Christopher
I'm not going to ask how you killed the MOBI but it sounds like a faster way than just writing to it

[B.A.T]

@Anvil

The Beta has expired and I had to do a restart trying to figure out what is wrong with SSDLife Pro. Is there an update coming soon?

Never mind. I found it in the ASU thread

[bulanula]

Too bad the mobi died because we really needed an SLC drive in here along with a C300 !

[Christopher]

Too bad the mobi died because we really needed an SLC drive in here along with a C300 !

Fear not. I'm on the case. And really, I can't think of a better SLC drive to test than a 16GB MOBI.. An Intel X25 E 32 could probably be done in a reasonable timeframe, but I prize my E drives. The MOBI can sustain fast enough writes to make testing feasible. I have a 32GB MOBi on the way, and I'm trying to get another 16GB. I have an MTRoN PRO 16GB to use if not.

[Christopher]

12 hours later and the Force 3 is not doing well

Looks like it died during a TRIM. (it did, the screenshot tells the story)

Will be updating the stats later tonight.

@Christopher
I'm not going to ask how you killed the MOBI but it sounds like a faster way than just writing to it

It's a short story. I shorted the drive out -- a nearly impossible task.

It was hot, even through the plastic chassis for minutes afterwards. The giant blue arc of electricity was difficult to miss.

Don't ask how I accomplished this -- but I've got more MTRONS to kill -- one way or the other.

[Hopalong X]

It's a short story. I shorted the drive out -- a nearly impossible task.

It was hot, even through the plastic chassis for minutes afterwards. The giant blue arc of electricity was difficult to miss.

Don't ask how I accomplished this -- but I've got more MTRONS to kill -- one way or the other.

I told you 120v AC would not make it write faster.

Next time try the Tesla Coil.

[B.A.T]

Todays update.
Kingston V+100
353.1491 TiB
2120 hours
Avg speed 25.37 MiB/s
AD still 1.
168= 1 (SATA PHY Error Count)
P/E?
MD5 OK.
Reallocated sectors : 00


It dropped out again but now it's running ok again.

Intel X25-M G1 80GB
289,78 TiB
20575 hours
Reallocated sectors : 00
MWI=124 to 121
MD5 =OK
51.09 MiB/s on avg


m4
322.9721 TiB
1200 hours
Avg speed 78.67 MiB/s.
AD gone from 180 to 167.
P/E 5693.
MD5 OK.
Reallocated sectors : 00


I'm still having problems with SSDLife Pro. We are testing out a lot of solutions but no luck yet. You guys just have to settle with some CDI screenshots.

[Anvil]

Kingston SSDNow 40GB (X25-V)

669.36TB Host writes
Reallocated sectors : 05 21
Available Reserved Space : E8 99
POH 6035
MD5 OK

32.80MiB/s on avg (~46 hours)

--

Corsair Force 3 120GB

Ended at
Host Writes 1015.n TiB

Will post the last few screenshots, unfortunately none showing CDI.

--

[Anvil]

Last (w)rites of the Force 3

Benchmark, performed minutes before it died


Last thing that happened


The caddy


close-up of SF and NAND

[B.A.T]

R.I.P Force 3