SSD Write Endurance 25nm Vs 34nm

[bulanula]

How long before the 320 is really dead with the slowing writes and all that ?

[B.A.T]

Todays update without screenshots. The forum just won't cooperate tonight. I'll put them in later in an edit.
m4
764.2669 TiB
2735 hours
Avg speed 88.25 MiB/s.
AD gone from 174 to 171.
P/E 13257.
Value 01 (raw read error rate) has changed from 30 to 43.
MD5 OK.
Reallocated sectors : 00


Kingston V+100:
185.1462 TiB
732 hours
Avg speed 75.81 MiB/s
AD gone from 17 to 8.
P/E?
MD5 OK.
Reallocated sectors : 00


Intel X25-M G1 80GB
13,7370 TiB Host writes
18696 hours
Reallocated sectors : 00
MWI=86 to 79
MD5 OK
52.08 MiB/s on avg


Filling the remaining space with a large file and deleting it worked like a charm. I'll just do that at the same time I update the screenshot

MTRON
Today I tried it on my work laptop without any luck. I get the same problems I had yesterday.

[Anvil]

Kingston SSDNow 40GB (X25-V)

428.58GB Host writes
Reallocated sectors : 12

MD5 OK

33.31MiB/s on avg (~95 hours)

--

Corsair Force 3 120GB

01 85/50 (Raw read error rate)
05 2 (Retired Block count)
B1 82 (Wear range delta)
E6 100 (Life curve status)
E7 34 (SSD Life left)
E9 251665 (Raw writes)
F1 335055 (Host writes)

MD5 OK

102.54MiB/s on avg (~95 hours)

power on hours : 1001

B1 (WRD) is down again, it's now at 82.

[Christopher]

I cant seem to upload any files either, so I'm going to post them when I can. I'm 450 miles away at the moment, but I've got screen captures so I'll handle that tomorrow... if XS cooperates.

[bulanula]

Any more news about the Samsung and the 320 drives ? Vapor ?

[Anvil]

Kingston SSDNow 40GB (X25-V)

430.00GB Host writes
Reallocated sectors : 12

MD5 OK

33.24MiB/s on avg (~108 hours)

--

Corsair Force 3 120GB

01 92/50 (Raw read error rate)
05 2 (Retired Block count)
B1 81 (Wear range delta)
E6 100 (Life curve status)
E7 33 (SSD Life left)
E9 255090 (Raw writes)
F1 339608 (Host writes)

MD5 OK

102.56MiB/s on avg (~108 hours)

power on hours : 1013

B1 (WRD) is down again, it's now at 81.

[Ao1]

When SSD Write Endurance testing is part of your job,

This is for one 25nm test:
Snip

Snip
(chamber temperature is 0-75C)

Any chance of some insights into why the B1 values are so different between the three SF drives currently being tested? SF drives are supposed to limit the delta between most and least worn blocks, yet even Christopher’s drive seems to be above the parameters that SF specify.

[Christopher]

Ao1, mine is at 9 now.

[bluestang]

M225->Vertex Turbo 64GB Update:

601.75 TiB (661.63 TB) total
1610.39 hours
12071 Raw Wear
117.32 MB/s avg for the last 17.49 hours (on W7 x64)
MD5 OK
C4-Erase Failure Block Count (Realloc Sectors) at 10.
(1=Bank 6/Block 2406; 2=Bank 3/Block 3925; 3=Bank 0/Block 1766; 4=Bank 0/Block 829; 5=Bank 4/Block 3191; 6=Bank 7/Block 937; 7=Bank 7/Block 1980; 8=Bank 7/Block 442; 9=Bank 7/Block 700; 10=Bank 2/Block 1066)

[Anvil]

Kingston SSDNow 40GB (X25-V)

431.31GB Host writes
Reallocated sectors : 12

MD5 OK

33.19MiB/s on avg (~120 hours)

--

Corsair Force 3 120GB

01 92/50 (Raw read error rate)
05 2 (Retired Block count)
B1 80 (Wear range delta)
E6 100 (Life curve status)
E7 32 (SSD Life left)
E9 258282 (Raw writes)
F1 343849 (Host writes)

MD5 OK

102.58MiB/s on avg (~120 hours)

power on hours : 1025

B1 (WRD) is down again, from 81 to 80.

[devsk]

Looks like Vertex Turbo is going to overtake M4.... This is a good race! An old timer and the new kid on the block.

[Anvil]

Kingston SSDNow 40GB (X25-V)

432.82GB Host writes
Reallocated sectors : 12

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 79 (Wear range delta)
E6 100 (Life curve status)
E7 31 (SSD Life left)
E9 261885 (Raw writes)
F1 348640 (Host writes)

MD5 OK

102.58MiB/s on avg (~133 hours)

power on hours : 1038

B1 (WRD) is down again, from 80 to 79.

[Ao1]

Updated MWI/B1 graph with a trend line added. Wear levelling between least and most worn blocks is so different its looks like three completely different drives.

@One_Hertz. Anything interesting from the post mortem?


http://imageshack.us/photo/my-images/546/unledaoa.png/

[Christopher]

Looks like Vertex Turbo is going to overtake M4.... This is a good race! An old timer and the new kid on the block.

That one drive has 25nm Micron and the other has 51nm Samsung is a little surprising. If they end up pulling close to each other at the end, I think some of the rules might not apply.

[Christopher]

Ao1,

I could try to add more static data to the Mushkin. Isn't the force 3 using 45GB of static data?

[devsk]

That one drive has 25nm Micron and the other has 51nm Samsung is a little surprising. If they end up pulling close to each other at the end, I think some of the rules might not apply.

no, I meant vertex writes faster (by about >20%) than M4 and still going. So, it has the chance to overtake M4. But realistically, I have a feeling that its gonna get tired and die before it overtakes M4. But we will see. That's the reason why this race is exciting...

[bluestang]

Only I can't run mine 24/7 like he can.

[Christopher]

Only I can't run mine 24/7 like he can.

Sadness!

I'm very happy with my new 24\7 system. I got the kinks out just in time to leave town for a couple days and have the confidence to just check in with remote apps. The whole system uses slightly more power than my cable box. Today is day 38, and I've averaged 9.6TB host, 7.5TB nand writes. Though I've lost several days time due to disconnects over the weeks, I'm sitting at 96 hours uptime!

[bluestang]

It's not cause there is any problems. I have a dual boot system at work...during the day (work time) I have to use XP for now and at night I let the M225 run ASU on W7 Ent x64. So it only get ~16 hrs/day, excpept for weekends were it's run on W7 all weekend.

[devsk]

It's not cause there is any problems. I have a dual boot system at work...during the day (work time) I have to use XP for now and at night I let the M225 run ASU on W7 Ent x64. So it only get ~16 hrs/day, excpept for weekends were it's run on W7 all weekend.

No wonder it stayed behind M4 despite it being much faster.

[B.A.T]

B.A.T
Fridays number. My rig has been down since friday afternoon and will be up again tomorrow. I moved it from work and back home.
I can still not upload pictures. Something is wrong with XS.
m4
768.5687 TiB
2749 hours
Avg speed 88.43 MiB/s.
AD gone from 171 to 169.
P/E 13330.
Value 01 (raw read error rate) has changed from 43 to 46.
MD5 OK.
Reallocated sectors : 00

Kingston V+100:
188.7754 TiB
746 hours
Avg speed 75.40 MiB/s
AD gone from 8 to 4.
P/E?
MD5 OK.
Reallocated sectors : 00

Intel X25-M G1 80GB
16,2689 TiB Host writes
18710 hours
Reallocated sectors : 00
MWI=79 to 75
MD5 OK
51.22 MiB/s on avg

MTRON
I will test it in my foldingrig tomorrow.

[Anvil]

Kingston SSDNow 40GB (X25-V)

435.55GB Host writes
Reallocated sectors : 12

MD5 OK

33.30MiB/s on avg (~156 hours)

--

Corsair Force 3 120GB

01 95/50 (Raw read error rate)
05 2 (Retired Block count)
B1 77 (Wear range delta)
E6 100 (Life curve status)
E7 29 (SSD Life left)
E9 268470 (Raw writes)
F1 357387 (Host writes)

MD5 OK

103.25MiB/s on avg (~156 hours)

power on hours : 1062

B1 (WRD) is down again, from 79 to 77.

It's been working for almost 1 week, so, a new record on the new firmware.

[B.A.T]

Since my wife has gone into labor I'll stop the test until I'm back from the hospital. I'll fire it up again when we are all back home.

[bluestang]

Congrats!

baby B.A.T.

[Ao1]

Here is a chart that compares the amount of writes incurred before the MWI expired. The SF stats are based on the assumption that when the MWI gets to 10 the theoretical P/E cycle capability has expired. Considering Christopher & SynbiosVyse used compressible data (46%?) it didn’t really provide an advantage. Without compression writes would be well below the 320 & M4.

Congrats B.A.T.

http://imageshack.us/photo/my-images/405/unledpsw.png/