SSD Write Endurance 25nm Vs 34nm

[Christopher]

I don't have access to my X25-E at the moment, and the only pics I have access to are some benchmarks.

But the power count/unsafe shutdown counts are not 1:1, but there are quite a few. Mine only has around 300GB on it though.

[Anvil]

I checked my X25-E drives and all have unsafe shutdowns, looks like 20-30% of the shutdowns are unsafe.

I performed a few tests last night and when Unsafe shutdown increased, the Power Cycle Count did not.
(I had expected Power Cycle Count to increase on every power Off and it didn't)

[Christopher]

I checked my X25-E drives and all have unsafe shutdowns, looks like 20-30% of the shutdowns are unsafe.

I performed a few tests last night and when Unsafe shutdown increased, the Power Cycle Count did not.
(I had expected Power Cycle Count to increase on every power Off and it didn't)

But at the same time, it seems as though most drives don't act this way... the Intel's power cycle may = 1 power on + 1 ATA shutdown. If the unsafe shutdown increases, power cycle doesn't. Other drives just assume that the drive has to be on to get turned off

So the true number of power cycles for an Intel is Power Cycle Count + Unsafe Shutdown.


I believe that if you enter the bios/uefi then exit, unsafe shutdown count will increase. I enter the BIOS and UEFI all the time, so some drives are worse off than others. For example, I had been reflashing and upgrading the FW on all of my Indilinx drives, so I would remove all of the other drives in the system, as it could take a lot of reboots and UEFI changes and I'd prefer not to have other drives in the system when doing FW updates. I don't think the unsafe shutdown counts are problematic, but it's probably best to avoid them if you can do so without much trouble. Basically, anything other than a soft-shutdown with a modern OS would result in an unsafe shutdown count.

All of the drive failures I've experienced have occurred in Windows, not on a restart. The sample size is pretty small though.

[minpayne]

I checked my X25-E drives and all have unsafe shutdowns, looks like 20-30% of the shutdowns are unsafe.

I performed a few tests last night and when Unsafe shutdown increased, the Power Cycle Count did not.
(I had expected Power Cycle Count to increase on every power Off and it didn't)

But at the same time, it seems as though most drives don't act this way... the Intel's power cycle may = 1 power on + 1 ATA shutdown. If the unsafe shutdown increases, power cycle doesn't. Other drives just assume that the drive has to be on to get turned off

So the true number of power cycles for an Intel is Power Cycle Count + Unsafe Shutdown.

Can I ask you guys what's the firmware of your X25-E's? Mine is 045C8850. This disk has been used in 4 different computers during the past 2 years and I don't know which computer(s) might did most of the "harm" to it.

I've done some simple tests of the "0C Power Cycle Count" and the "C0 Unsafe Shutdown Count" here in the current computer (ThinkPad X220, BIOS 1.25):

1474
1033

soft shutdown

1475
1034

soft reboot

1475
1034

sleep

1476
1035

hibernate

1477
1036

It looks like the difference between my 0C and C0 is locked I'll test it in another computer and see if it's the same problem.

Edit: just confirmed that my desktop E760 motherboard doesn't increase the "C0 Unsafe Shutdown Count" when doing soft shutdowns in Windows. It's obviously something wrong with the ThinkPad X220.

[minpayne]

Strange, now I cannot reproduce any problem on the 3rd sample of the M4 512GB. Today when I woke up it was the first time I saw the MWI lost virginity (no reallocated sector though, strange? Is it because MWI is calculated according to wear-leveling count and the P/E of the flash?):

[minpayne]

OK never mind, I've figured it out myself. By looking into the figures in the OP of this thread, it is very obvious that the MWI is simply calculated by the P/E spec of the flash. For M4, the P/E spec is 3000 cycles, and so the MWI of the M4 64GB should last for 59.5GB * 3000 / 1024 = 174.3 TB, which is pretty close to that of the test result of the first M4 64GB sample in this thread (and also the trend of the 2nd sample). It also works the same for the Samsung 470 64GB. I'm just a bit slow to follow you guys here

[Christopher]

OK never mind, I've figured it out myself. By looking into the figures in the OP of this thread, it is very obvious that the MWI is simply calculated by the P/E spec of the flash. For M4, the P/E spec is 3000 cycles, and so the MWI of the M4 64GB should last for 59.5GB * 3000 / 1024 = 174.3 TB, which is pretty close to that of the test result of the first M4 64GB sample in this thread. It also works the same for the Samsung 470 64GB. I'm just a bit slow to follow you guys here

I've not had the opportunity to observe my M4 decrease MWI, but other drives operate differently. Also, the NAND in the M4 is rated at 3000PE cycles, but it really should be rated at 5000.

An Indilinx drive typically has high write amplification with random data, so MWI on a 120GB Turbo may drop after 100MB of host writes (when used as a OS drive, sometimes less). So running ASU will typically take much more host writes to do reduce MWI by 1. Actually, MWI with Indilinxes are kind of wonky anyway so perhaps they're not the best example.

I just opened a brand new 64GB Vertex Turbo last night, and it was at 97 percent out of the box (1.7FW). A brand-new Vertex EX 60 was at 89 MWI out of the box when I unsealed it last week (1.3FW).

The 64GB "Real Turbo" is not half as awesome as Bluestang's M225 Turbo, at least not on 1.7FW. It also had 5 erase failures out of the box too.

That doesn't mean the MWI is a good indicator of anything in and of itself. When taken in context with other factors, it can help give you a picture of what is happening -- but its just part of the picture and doesn't mean much on it's own.

[minpayne]

I just noticed that the MWI of Intel 320 Series may also work differently.

Did some simple tests of my Intel 600GB drive and yet to observe the MWI increase (I thought the MWI was supposed to increase by 1 when 5.5TB is written, if the P/E spec is 1000?):





Maybe the wear-leveling algorithm of the Intel 320 Series isn't that good, and without over-provisioning (which means merely relying on the factory reserved space) the drive could wear out pretty fast on some blocks?

[minpayne]

That doesn't mean the MWI is a good indicator of anything in and of itself. When taken in context with other factors, it can help give you a picture of what is happening -- but its just part of the picture and doesn't mean much on it's own.

Then what are we looking for in this thread? When the MWI is exhausted on a drive, the flash becomes volatile to some extent. How to verify if it's still working as it's supposed to? (It's obviously not practical to wait for a year of loss of power then verify the integrity of data in it.) A drive may continue to work to 50% more of its spec'ed P/E cycles but how do you know the current period that the data can be held without power?

[Anvil]

The 320 Series 600GB would need about 17.5TiB written for every decrease of MWI if P/E is at 3000.

E9 starts at 100 and stops decreasing at 1.

There's nothing wrong with Intel's reporting , it's just a huge drive.

--

edit

All my E's except for one are 8850, the last one has 8860 which is not available as a download.

[B.A.T]

Todays update:
Kingston V+100
310.3283 TiB
1597 hours
Avg speed 25.36 MiB/s
AD still 1.
168= 1 (SATA PHY Error Count)
P/E?
MD5 OK.
Reallocated sectors : 00


Intel X25-M G1 80GB
167,5932 TiB
19815 hours
Reallocated sectors : 00
MWI=155 to 154.
MD5 =OK
52.75 MiB/s on avg


m4
119.5655 TiB
432 hours
Avg speed 81.20 MiB/s.
AD gone from 34 to 31.
P/E 2083.
MD5 OK.
Reallocated sectors : 00

[Anvil]

Kingston SSDNow 40GB (X25-V)

577.73TB Host writes
Reallocated sectors : 05 14
Available Reserved Space : E8 99
POH 5231
MD5 OK

33.15MiB/s on avg (~133 hours)

--

Corsair Force 3 120GB

01 92/50 (Raw read error rate)
05 2 (Retired Block count)
B1 48 (Wear range delta)
E6 100 (Life curve status)
E7 10 (SSD Life left)
E9 572928 (Raw writes) ->560TiB
F1 762590 (Host writes) ->745TiB

MD5 OK

106.68MiB/s on avg (~58 hours)

power on hours : 2212

[Anvil]

Chart on the Force 3, showing WRD vs GiB written over the last few days.



It shows that WRD climbs rapidly vs the GiB needed to decrease WRD to the same level.

[Christopher]

Nice chart Anvil.

Are you using logged attributes or doing it manually?

[B.A.T]

Christmas update:
Kingston V+100
311.8668 TiB
1614 hours
Avg speed 25.35 MiB/s
AD still 1.
168= 1 (SATA PHY Error Count)
P/E?
MD5 OK.
Reallocated sectors : 00


Intel X25-M G1 80GB
170,8411 TiB
19833hours
Reallocated sectors : 00
MWI=154
MD5 =OK
52.80 MiB/s on avg


m4
124.5444 TiB
449 hours
Avg speed 81.21 MiB/s.
AD gone from 31 to 28.
P/E 2169.
MD5 OK.
Reallocated sectors : 00


My next update will be late tomorrow night. No it's time to enjoy christmas with my family.

Merry Christmas to you all

[Anvil]

Nice chart Anvil.
Are you using logged attributes or doing it manually?

I'm using logged attributes

Christmas update:

Merry Christmas to you all

Merry Christmas to all of you!

[bluestang]

Yeah, Merry Christmas and Happy New Year to all!!!

[Computurd]

merry christmas~!

[devsk]

Merry Christmas to you all!

[minpayne]

I bought myself a Christmas present - Samsung 830 512GB, and I wish everyone here a happy new year

[B.A.T]

Christmas update:
Kingston V+100
313.9284 TiB
1638 hours
Avg speed 25.33 MiB/s
AD still 1.
168= 1 (SATA PHY Error Count)
P/E?
MD5 OK.
Reallocated sectors : 00


Intel X25-M G1 80GB
175,1643 TiB
19857 hours
Reallocated sectors : 00
MWI=154 to 152
MD5 =OK
52.68 MiB/s on avg


m4
131.2455 TiB
473 hours
Avg speed 81.22 MiB/s.
AD gone from 28 to 24.
P/E 2286.
MD5 OK.
Reallocated sectors : 00

[Anvil]

Kingston SSDNow 40GB (X25-V)

583.01TB Host writes
Reallocated sectors : 05 15 Increased by 1
Available Reserved Space : E8 99
POH 5277
MD5 OK

34.18MiB/s on avg (~20 hours)

--

Corsair Force 3 120GB

01 95/50 (Raw read error rate)
05 2 (Retired Block count)
B1 49 (Wear range delta)
E6 100 (Life curve status)
E7 10 (SSD Life left)
E9 585940 (Raw writes) ->572TiB
F1 779900 (Host writes) ->762TiB

MD5 OK

106.87MiB/s on avg (~20 hours)

power on hours : 2258

--

I had to update a utility and so the testing on both drives were re-started about 20 hours ago.

[Anvil]

An update on WRD and then a new one that includes Raw Error Read Rate for the Force 3

[Anvil]

A brief summary so far.





...

[Christopher]

Christmas will be over here in seven minutes, but I thought I would wish you all a merry Christmas before it is over.

Well then, let me be the first to wish you a happy new year. Hopefully the Samsung will die soon as the testing rig is going to start getting crowded in 2012.