SSD Write Endurance 25nm Vs 34nm

[Kain665]

oh lord Your feisty, and I like that but then again, so was my last cellmate, and we all know how that ended......

[DragoonXX]

If you budget stretches enough, the Samsung 830 is a great choice. Fast and reliable.

No-one really knows how long you will get out of a drive, that is pretty random .... what I do know is that you are not going to wear it out doing any normal sort of things with your computer any time soon.

I can echo the -not going to wear it out doing any normal sort of things with your computer anytime soon.- bit. when I first got my SSDs, I too was under the impression of-extremely fragile/shortlived and followed every SSD tweak guide I could find, which wasn't really much of an issue with me because I have multiple drives to store data to, and got used to it very quickly. My 2 (seemingly unkillable from the testing happening here) X-25V drives stand right at this moment with:
Power on Hours: 14 929 and 14 978
MWI:99 and 99
Host Writes:1.56TB and 1.70TB
as reported by the intel SSD toolbox (in a RAID0 array- where they have spent ~95% of their life so far)

[B.A.T]

Todays update:

Intel X25-M G1 80GB
537.63 TiB
22529 hours
Reallocated sectors : 114 to 120
Available Reserved space: 93 to 92
MWI=176 to 169
MD5 =OK
43.06 MiB/s on avg


m4
748.1641 TiB
3195 hours
Avg speed 73.08 MiB/s.
AD gone from 185 to 181.
P/E 12938.
MD5 OK.
Reallocated sectors : 00

[Christopher]

Samsung 830 256 Day 36

846,887.35 GiB
827.03 TiB

290.36 MB/s Average
842 Hours

3691 Wear Leveling Count
MWI 1

0/0 Erase/Program Fail

OCZ Vertex Turbo 64 Day 53

396,957.33 GiB
387.65 GiB
84.60 MB/s Avg
1340 hours

7096 Avg PE count
-100 MWI

0/0/0 Read/Program/Erase Fail

--------------------------------------------------\\
The Samsung 830 finally hit MWI 1 at 1/1/3600 WLC 177.
I'll look through the logs to try and estimate when it hit MWI 1 more precisely.
--------------------------------------------------\\
I'm about to perform some testing on the Octane to see if I can resurect it.

Wish me luck.

[canthearu]

You have mixed up the TiB written for the Turbo and the 830

[Christopher]

So I took some time to see if I could resurrect the Octane.

Turns out, it wasn't necessary.

THE OCTANE LIVES!!

I'd very much like to take credit for this, that through my vast powers I was able to bring the dead back to life... but it is not so. It just crapped out and needed a power cycle. That is not very desirable behavior from a modern drive, but it's not like it happened once every 30 hours.

Apparently, it just pulled a [insert your own reference here] on me. It just... dropped out. Since I wasn't there to physically confirm this until a short while ago, testing is back on.

[Christopher]

Other business --

OCZ Octane 128GB--
On the power cycle, POH reset but 6 new blocks went bad (while it was sleeping presumably).

OCZ Vertex Turbo 64GB --
Oh, and 1 of the 3 "Indilinx-Specific" SMART attributes have increased from 0 to 13 (attribute D3). I have no idea what this means, but I'd very much like to know.

Samsung 830 256GB --
The 830s CRC Error Count has increased to 2 (attribute C7)


-----------------
Actually, it looks like the Indilinx D3 attribute is SATA Error Count CRC. If the 830 and the VT are having the same problem at the same time, then it could be a host issue (??).

[B.A.T]

You just experienced the same as I did with the Kingston V+100, and it sucks every time it happens.

[Ao1]

You just experienced the same as I did with the Kingston V+100, and it sucks every time it happens.

Is the Kingston V+ using a Samsung controller or a Toshiba controller?

Product # SNV225 = Samsung
Product # SNVP325 = Toshiba

[B.A.T]

The Kingston V+100 uses an T6UG1XBG from Toshiba.

[canthearu]

Patriot Torqx-2 64GB - Day 67

Drive hours: 1,618
ASU GiB written: 109,693.08 GiB (107.12 TiB)
Avg MB/s: 25.43 MB/s
MD5: OK

Bad blocks: 2 grown, 83 factory
Wear cycle counter: 6872 average erase cycles, 10502 maximum erase cycles.
Total ECC Error Count: 1

Intel 520 60GB - Day 59

Drive hours: 1,416
ASU GiB written: 442,356.84 GiB (431.99 TiB)
Avg MB/s: 95.60 MB/s
MD5: OK

Host GB written (F1): 445,423.94 GiB (434.98 TiB, 14253566 raw)
NAND writes (F9): 315,361 GiB (307.97 TiB)

Reallocated sectors (05): 0
Failure count (AB, AC): 0 program, 0 erase
Raw Error Rate (8B): 117 normalised
Avaliable Reserved Space (AA): 100 normalised
Media Wearout Indicator (E9): 1

Vertex 4 128GB - Day 6

Drive hours: 160
ASU GiB written: 60,235.34 GiB (58.82 TiB)
Avg MB/s: 146.09 MB/s
MD5: OK

Sector GB written (E8): 61,285.02 GiB (59.85 TiB, 128523997843 raw)

Reallocated sectors (05): 0
Raw Read Error Rate (01): 6
Remaining Life (E9): 93

Note: New Endurance rig appears to be very stable. Not a single crash since starting endurance testing on this motherboard.

[B.A.T]

Todays update:

Intel X25-M G1 80GB
541.00 TiB
22552 hours
Reallocated sectors : 120 to 123
Available Reserved space: 92
MWI=169 to 163
MD5 =OK
43.02 MiB/s on avg


m4
753.8942 TiB
3218 hours
Avg speed 73.09 MiB/s.
AD gone from 181 to 178.
P/E 13033.
MD5 OK.
Reallocated sectors : 00


Only 3 days left

[Christopher]

The Octane... (sigh.) I'm not sure that it dropped out. It seems to have crashed the three programs that make use of SMART info and it's instance of ASU. I had to shut down the rig after months and months of uptime. It's put like 1.5PiB though the controller since it's last reboot.

I have moved the Octane to another rig temporarily to see what the hell is going on.

[B.A.T]

That is very strange. It probably is not your rig but your are right about testing it in a nother rig to make sure.

[Christopher]

Hmmm. I think the Octane is causing some BSoDs in my main system where it's now located. It's hard to be sure, but it seems like it's creating all sorts of random issues. I don't want to lay it all at the feet of the Octane, but it's looking like it.

Reallocations are up to 238 now, but MWI is still at 85.

-----------------------------------------------------
OCZ Octane 128GB Day ??

311,111.64 GiB
303.82 TiB

159.22 MB/s

Reallocations 238

-----------------------------------------------------
OCZ Vertex Turbo Day 55

402,345.40 GiB
392.91 TiB

84.24 MB/s

Avg Erase Count 7193

Program/Erase/Read Failures 0/0/0

1359 Hours

-----------------------------------------------------
Samsung 830 Day 37

866,754.41 GiB
846.45 TiB

288.07 MB/s

Wear Leveling Count: 1/1/3777

Program/Erase Failures 0/0

862 Hours

-----------------------------------------------------

The Endurance Rig was restarted about 2.5 hours ago.

[Computurd]

very interesting, what are your thoughts on the octane, rapidly dying?

[Christopher]

very interesting, what are your thoughts on the octane, rapidly dying?

God, I sure hope so.

I put it back in the e-rig ; It's crapped out again. I think everytime it hits a bad block it's gonna drop out.

And that's getting old already.

------------------------------------

Okay. The Octane is really, really dead now. It dropped out for the last time, and is now a lifeless hulk.

[DragoonXX]

Okay. The Octane is really, really dead now. It dropped out for the last time, and is now a lifeless hulk.

not even worth doing an autopsy or giving it final rights before burning it in a funeral pyre?

[Christopher]

not even worth doing an autopsy or giving it final rights before burning it in a funeral pyre?

I need to think about it for a while. What can I really say about the Octane?

"It worked great for 18 days, except for the SMART data."

Doesn't seem like the best eulogy.

[canthearu]

"It worked great for 18 days, except for the SMART data."

Not a great omen for my vertex 4. I hope it's improved ECC stops this sort of issue killing it so soon.

[Christopher]

Not a great omen for my vertex 4. I hope it's improved ECC stops this sort of issue killing it so soon.

It probably a matter of luck more than anything. If you get a drive with good flash, then great. If you get one with a piece of questionable flash, your mileage may vary... Significantly. The Vertex 4 has both 128bit ECC and NAND redundancy.

[Ao1]

According to Anandtech it doesn't:

"The Vertex 4 doesn't have OCZ's redundant NAND technology enabled, although the enterprise version of the drive (Intrepid 3) will likely turn it on."

[Christopher]

According to Anandtech it doesn't:

"The Vertex 4 doesn't have OCZ's redundant NAND technology enabled, although the enterprise version of the drive (Intrepid 3) will likely turn it on."

Well I'll be damned.

Also, Intrepid 3 is a fantastic name IMHO.

[canthearu]

Patriot Torqx-2 64GB - Day 68

Drive hours: 1,644
ASU GiB written: 111,958.77 GiB (109.33 TiB)
Avg MB/s: 25.34 MB/s
MD5: OK

Bad blocks: 2 grown, 83 factory
Wear cycle counter: 7011 average erase cycles, 10717 maximum erase cycles.
Total ECC Error Count: 1

Intel 520 60GB - Day 60

Drive hours: 1,441
ASU GiB written: 450,890.08 GiB (440.32 TiB)
Avg MB/s: 95.93 MB/s
MD5: OK

Host GB written (F1): 453,999.69 GiB (443.36 TiB, 14527990 raw)
NAND writes (F9): 321,432 GiB (313.9 TiB)

Reallocated sectors (05): 0
Failure count (AB, AC): 0 program, 0 erase
Raw Error Rate (8B): 118 normalised
Avaliable Reserved Space (AA): 100 normalised
Media Wearout Indicator (E9): 1

Vertex 4 128GB - Day 7

Drive hours: 187
ASU GiB written: 73,311.22 GiB (71.59 TiB)
Avg MB/s: 147.38 MB/s
MD5: OK

Sector GB written (E8): 74,426.03 GiB (72.68 TiB, 156082695233 raw)
Reallocated sectors (05): 0
Raw Read Error Rate (01): 6
Remaining Life (E9): 91

[Christopher]

-----------------------------------------------------
OCZ Vertex Turbo Day 56

407,417.01 GiB
397.86 TiB

80.48 MB/s

Avg Erase Count 7286

Program/Erase/Read Failures 0/0/0

1377 Hours

-----------------------------------------------------
Samsung 830 Day 38

885,280.76 GiB
864.53 TiB

294.50(!) MB/s

Wear Leveling Count: 1/1/3857

Program/Erase Failures 0/0

880 Hours

-----------------------------------------------------

The 830 should hit 1PiB next Tuesday.

And the Octane? Still dead.