SSD Write Endurance 25nm Vs 34nm

[bulanula]

Would be good to know what the drive was.

[Hopalong X]

@also...this quasi-failure with the intel has just really impressed me. much more impressed with that drive now than i would have been had there not been a failure!

Like an old Timex. "They take a licking and keep on ticking".

I would have to admit the Timex salt water torture test might be a bit extreme for the Intel.

[SynbiosVyse]

286.11 hours
168.0275 TiB written
40.94 MB/s
MD5 ok

05: 0
B1: 83
E7: 25%
E9: 119616
EA/F1: 172864
F2: 256

[B.A.T]

m4 update:
518.3935 TiB
1688 hours
Avg speed 91.04 MiB/s.
AD gone from 58 to 54.
P/E 9085.
MD5 OK.
Still no reallocated sectors


Kingston V+100
I'm still trying to figure out why it drops out, so the test is still halted.

[bulanula]

Really impressive to see the Crucial M4 getting above 500 TB. Will see if it really beats the C300 etc. but I highly doubt it because the controllers are pretty much the same and the 34nm NAND is much better etc.

[Anvil]

315.36TB Host writes
Reallocated sectors : 6
MD5 OK

32.93MiB/s on avg (49+ hours)

[jcool]

Speaking of the X25-E... here's my 32GB one with the most host writes. MWI of 90 still at 580TB written:

[Ao1]

Speaking of the X25-E... here's my 32GB one with the most host writes. MWI of 90 still at 580TB written:

Nice. That unsafe shutdown count looks a bit severe though. Good job it’s not a 320

[bluestang]

M225->Vertex Turbo 64GB Update:

296.48 TiB (325.98 TB) total
882.9 hours
6475 Raw Wear
118.84 MB/s avg for the last 111.42 hours (on W7 x64)
MD5 OK
C4-Erase Failure Block Count (Realloc Sectors) at 4
(Bank 6/Block 2406; Bank 3/Block 3925; Bank 0/Block 1766; Bank 0/Block 829)

[bulanula]

WOW it really is impressive to see how much better SLC is compared to MLC ! The Intel X25-E lacks TRIM but has a WA of 1.1 and 50nm SLC NAND. A M4 has TRIM and WA of 1.1 and 34nm MLC NAND. Sandforce often has actual 0.6 WA which is very nice and TRIM too etc. Capacity also plays a role here. For example a 64GB X25-E will last twice as long as a 32GB X25-E !

Imagine the almost ideal scenario for SSD endurance ( which does not exist currently but the technology is there ) : 90nm SLC NAND, 256GB capacity ( tons of reserve space too ), WA of 0.6 with compression controller like Sandforce and TRIM too etc.

[sergiu]

Speaking of the X25-E... here's my 32GB one with the most host writes. MWI of 90 still at 580TB written:

Just out of curiosity, can you detail a little (if allowed) what kind of load do you have on that SSD? I am asking because on normal desktop usage and light servers I saw no more than 5-10TB of data per year until now.

[jcool]

I can't.. bought it like this - just got in today... I wondered WTF the guy did with it as well.

[Vapor]

That X25-E averaged ~31.7MiB/sec for ~7.5 months in real usage

[B.A.T]

m4 update:
526.6830 TiB
1714 hours
Avg speed 91.06 MiB/s.
AD gone from 54 to 49.
P/E 9227.
MD5 OK.
Still no reallocated sectors


Kingston V+100
I'm still trying to figure out why it drops out, so the test is still halted.

[Computurd]

wow looks like a server drive, bet it was used in a caching environment. Cache disks can take a beating, but damn that is alot of resets for a server scenario...interesting, a good mystery!

[squawker]

Speaking of the X25-E... here's my 32GB one with the most host writes. MWI of 90 still at 580TB written

Just out of curiosity, can you detail a little (if allowed) what kind of load do you have on that SSD? I am asking because on normal desktop usage and light servers I saw no more than 5-10TB of data per year until now.

I can't.. bought it like this - just got in today... I wondered WTF the guy did with it as well.

That X25-E averaged ~31.7MiB/sec for ~7.5 months in real usage

wow looks like a server drive, bet it was used in a caching environment. Cache disks can take a beating, but damn that is alot of resets for a server scenario...interesting, a good mystery!

i'm following this thread with great interest, admiring the great job done by all participants and learning a lot.

With risk of being obvious i'd like to remember that with respect to Intel Toolbox Smart Attributes, some of them seem not to be very reliable, since there are old complaints about them in Intel SSD's Forum, like this one:

[QUOTENov 16, 2010 4:34 PM
intel G1 ssd reports weird unsafe shutdown cont][/QUOTE]

With me, one new G2, just after OS installation, toolbox was showing about 2 or 3 TB of host writes.

Keep up the good job.

[[XC] Oj101]

For example a 64GB X25-E will last twice as long as a 32GB X25-E !

If the 64GB is less than twice as fast as the 32GB it should last more than twice as long

[bulanula]

Def. looks like your X25-E was used in an enterprise environment probably as a ZFS cache disk or for database work.

[bluestang]

The drive isn't even on the manufacturer's web site, isn't listed or is listed with a "usually ships in 2-4 weeks" at the few etailers that carry the series -- and I think the etailer that I placed the order with doesn't really have them. So I put in the order anyway, and I'll just have to wait and see. That's why I didn't want to mention it. I'm actually incredibly excited about this particular drive. I probably shouldn't have even mentioned it, but I couldn't help myself.

I think I might be addicted.

I will curl up into the fetal position and cry like a little baby if they don't have this drive sitting in their warehouse in West Nowhere, just waiting to get on the truck.

Alright Christopher...Time for an update on the status of that SSD.
Order confirmed?, got a tracking number yet?

At least tease us with the Brand

[bulanula]

Looks like he won't budge. Why so serious ?

[B.A.T]

m4 update:
532.2283 TiB
1731 hours
Avg speed 91.09 MiB/s.
AD gone from 49 to 46.
P/E 9326.
MD5 OK.
Still no reallocated sectors


Kingston V+100
Still not startet the test again. I'll continue to find whats wrong.

Next update will be 26 Sept. I'm going away on a job trip. If the drive fails when i'm away I'll get a timestamp from ASU when it happened.

Never mind, I'm using teamviewer and will still update every day (after tip from Anvil )

[johnw]

If the 64GB is less than twice as fast as the 32GB it should last more than twice as long

No. If it writes "less than twice as fast" while being double the capacity, it should last LESS than twice as long...assuming it writes faster than the lower capacity model. If both models write at the same speed, then double the capacity should be about double the longevity (although free space and overprovisioning can make a difference).

[Vapor]

C300 Update

367.1TiB host writes, 1 MWI, 6190 raw wear indicator, 2048/1 reallocations, 63.05MiB/sec, MD5 OK




SF-1200 nLTT Update

225.75TiB host writes, 164.688TiB NAND writes, 20 MWI, 2635 raw wear (equiv), wear range delta 3, 56.2MiB/sec, MD5 OK

[zads]

WOW it really is impressive to see how much better SLC is compared to MLC !
...

As a reference point, at zero P/E cycles:
256GB 25nm consumer MLC drive has about ~20,000,000 NAND pages average with at least 1 (flipped) bit error on it...
256GB 25nm enterprise SLC drive has less than ~20 NAND pages with at least 1 bit error on it...

[bulanula]

I think the M4 will die soonish now.