 koitsuPremium,MVM
join:2002-07-16
Mountain View, CA
kudos:19 | reply to trparky
Re: Buggy Intel SSD 520 Series SSD Firmware said by trparky:Did you read my post about how the failure of the SSD seemed to coincide with an issue of the TRIM command to the SSD?
No, that comment hadn't shown up by the time I wrote my explanation. This just brings even more nonsense into the picture:
1. How are you issuing "TRIM commands" for the drive on a nightly basis? On Windows Vista onward you should not need to do that (presumably you're running the "Intel SSD Optimizer") using Intel's SSD Toolbox; that's intended for XP systems which do not offer native OS-level TRIM (see below). You should not be doing this daily unless you are doing massive amounts of I/O every day (I'm talking maybe 60-80GBytes of writes and deletes, daily, on a 100-120GB SSD).
2. Why are you doing this on Windows 8? Since Windows Vista the OS has had TRIM capability natively within the ATA I/O subsystem driver. The OS takes care of this natively, and cleanly, for you per every I/O delete operation (or may submit large consecutive/linear LBA blocks in an optimal way, rather than one at a time -- both methods are fine). You should let the OS take care of this for you; it will do a more efficient job and consistently 100% of the time.
3. Are your partitions on your SSD properly aligned to either 1MByte or 2MByte boundaries (or possibly other multiples of (2^10)*4)? Windows Vista onward has ensured that, but if you did something like use XP back in the day, then installed Vista (including a clean install but without deleting the existing partition (not the same as formatting the existing partition; that would not change the alignment)) or newer, then that would explain very bad TRIM performance in general. I'm not talking about 4KByte alignment here, I'm talking about NAND erase block size alignment (which is represented by N number of sequential NAND pages, and the NAND page size varies per SSD brand, model/device, revision, and lithography. Most manufacturers do not disclose this info, which still to this day pisses me off). You can read about the ill effects of non-NAND-erase-block-aligned partitions here (read, do not skim): »wiki.laptop.org/go/How_to_Damage···e_Device
4. How much free space did your Intel SSD have before you deleted 10GB of data off of it, and what capacity is that SSD? You should always keep roughly 30% of free space on the SSD.
5. Have you ever done something like a "full format" on the SSD, i.e. every LBA ("sector") written to with data (or zeros)? If so, this would explain awful performance of the SSD, especially during TRIM or GC operations -- the FTL is completely maxxed out. Do not do this on an SSD. ATA-level Secure Erase is the proper way to do this.
6. Did you ever look at any of the SMART statistics on your Intel SSD? If so, do you have that data somewhere (screenshot, etc.)? It would give me some indication of its internal state.
TRIM in general is an expensive operation, and likewise, GC (garbage collection) takes even longer. The drive can go catatonic during this state, to the point where kernel/device drivers may think the underlying device has "fallen off the bus" (in actuality the kernel/drivers hits an internal I/O timeout and then kicks the drive off the bus itself). Any OS will see this. The timeout on FreeBSD is 30 seconds with a 5-attempt retry count. What Windows uses as a timeout depends on the underlying storage drivers; if you're using Microsoft's AHCI, you would need to ask them or look through MSDN. If you're using Intel's RST drivers (be sure to state what version) you would need to ask Intel what the value is.
--
Making life hard for others since 1977.
I speak for myself and not my employer/affiliates of my employer. |
|
| reply to trparky
Sh*t. I have two 520 SSDs. Guess I'll just cross my fingers. Wonder if they'll ever release a fixed firmware? I thought that Intel's firmware was supposed to be free of the SandForce plague? I know it's the same controller but I thought that the firmware was different. |
|
 Hyrules
join:2006-07-19
Gatineau, QC
kudos:1
1 edit | reply to trparky
It's sad that so many SSD have problems. Intel has problems with theirs, corsair as well. Technology is too new.
--
- Technicien en informatique / Computer tech [A+]
- Chasseur d'orage / Storm Chaser |
|
 DarkLogixTexan and ProudPremium
join:2008-10-23
Baytown, TX
kudos:3 | reply to koitsu
This is why I always set windows to not turn off the harddrive ever.
though oddly and I blame marvell my intel 520 480gb shows a crazy number of unclean shutdowns.
But my 520 is running great.
--
»Death Star Petition |
|
|
|
 rusdiAmerican VPremium,MVM
join:2001-04-28
Flippin, AR
kudos:1
1 edit | reply to trparky
The only trouble I have had with mine, is slow boot.
I'm now curious if this might be affecting your Intel drive as well. Maybe on a different level.
If you're willing to make a few Registry changes:
»Slow boot time on your SSD?
This is for Windows 7. |
|
 trparkyApple... YUMPremium,MVM
join:2000-05-24
Cleveland, OH
kudos:2
 ·Time Warner Cable
| I thought that TRIM was supposed to be done even if Windows does it itself. I know that Windows performs a TRIM when you delete files but what about when you overwrite a file?
Microsoft hasn't exactly been very forthcoming with their policy on the TRIM command that's built into the OS. When does a TRIM happen?
--
Tom
Boycott AT&T uVerse! | Tom's Android Blog | AOKP (The Android Open Kang Project) |
|
 GhastlyonePremium
join:2009-01-07
Las Vegas, NV
kudos:2 | reply to DarkLogix
said by DarkLogix:This is why I always set windows to not turn off the harddrive ever.
How do you do that?
I must have overlooked that option in there. |
|
 CheesePremium
join:2003-10-26
Naples, FL
kudos:1 | It's in power options |
|
rusdiAmerican VPremium,MVM
join:2001-04-28
Flippin, AR
kudos:1 | reply to trparky
said by trparky:I thought that TRIM was supposed to be done even if Windows does it itself. I know that Windows performs a TRIM when you delete files but what about when you overwrite a file?
Microsoft hasn't exactly been very forthcoming with their policy on the TRIM command that's built into the OS. When does a TRIM happen?
It "should" be.
Here's a way to check if it's enabled.
Open a command prompt, (Administrative level).
Command prompt > fsutil behavior query disabledeletenotify
DisableDeleteNotify = 1 (Windows TRIM commands are disabled)
DisableDeleteNotify = 0 (Windows TRIM commands are enabled) |
|
trparkyApple... YUMPremium,MVM
join:2000-05-24
Cleveland, OH
kudos:2 | I know about that, but appears that that may indicate that a TRIM only happens when you delete files. What about an overwrite of a file? |
|
rusdiAmerican VPremium,MVM
join:2001-04-28
Flippin, AR
kudos:1 | said by trparky:I know about that, but appears that that may indicate that a TRIM only happens when you delete files. What about an overwrite of a file?
Overwrite doesn't happen with NAND flash. it must be deleted and written in other blocks...
Here's a nice article that might help explain the difference between TRIM, and Garbage collection.
»thessdreview.com/daily-news/late···-primer/ |
|
DarkLogixTexan and ProudPremium
join:2008-10-23
Baytown, TX
kudos:3 | reply to Ghastlyone
See here. |
|
DarkLogixTexan and ProudPremium
join:2008-10-23
Baytown, TX
kudos:3 | reply to trparky
said by trparky:I know about that, but appears that that may indicate that a TRIM only happens when you delete files. What about an overwrite of a file?
Due to wear leveling you don't ever overwrite
overwriting nand is very slow and time consuming so instead it writes to a new location and marks the old for GC to clean up
--
»Death Star Petition |
|
GhastlyonePremium
join:2009-01-07
Las Vegas, NV
kudos:2 | reply to DarkLogix
Awesome, thanks.
So this doesn't effect sleep mode, or shutting down completely? |
|
DarkLogixTexan and ProudPremium
join:2008-10-23
Baytown, TX
kudos:3 | If you shut down completely it'll still turn it off, but the drive won't (in the case of MHDD's) spin down. while the computer is on.
I think putting the computer in sleep mode would still turn it off.
I always set this setting because with a MHDD I don't want to have to wait on spin up to use my computer again, and I figure for a SSD it'd be more ideal for the drive to be as ready as possible.
Though I always kinda figured that a SSD wouldn't really do much when windows tells it to "spin down"
--
»Death Star Petition |
|
Hyrules
join:2006-07-19
Gatineau, QC
kudos:1 Reviews:
 ·FreePhoneLine
·Videotron
 ·ELECTRONICBOX
1 edit | reply to trparky
disabling hdd power down in windows is the first thing I do after I reinstall my computer. I does not really help even with SSD. It might avoid some problems but in this case it doesn't with my pc.
--
- Technicien en informatique / Computer tech [A+]
- Chasseur d'orage / Storm Chaser |
|
GhastlyonePremium
join:2009-01-07
Las Vegas, NV
kudos:2 | reply to trparky
Ok good to know. I've got my PC set to High Performance. So it's probably already disabled. |
|
DarkLogixTexan and ProudPremium
join:2008-10-23
Baytown, TX
kudos:3 | said by Ghastlyone:Ok good to know. I've got my PC set to High Performance. So it's probably already disabled.
I'd double check to be sure.
--
»Death Star Petition |
|
koitsuPremium,MVM
join:2002-07-16
Mountain View, CA
kudos:19 | reply to trparky
said by trparky:I thought that TRIM was supposed to be done even if Windows does it itself. No no no. Absolutely not!
said by trparky:I know that Windows performs a TRIM when you delete files but what about when you overwrite a file?
Microsoft hasn't exactly been very forthcoming with their policy on the TRIM command that's built into the OS. When does a TRIM happen?
The answer lies within truly understanding what TRIM does. The entire point of the TRIM command (more specifically DATA SET MANAGEMENT; that's what it's called per ATA spec) is to allow an OS or host controller to tell the underlying SSD "these LBA ranges are no longer used". Meaning: the entries in the FTL map can be released, and the SSD internally knows anything from LBA x to LBA y is no longer cared about.
Obviously this is used during file deletions; the OS deletes the file from the underlying filesystem ("I no longer care about this file"), and issues the appropriate TRIM command to the controller saying "LBA x through y are no longer used" (and possibly in repeated chunks, since that ATA CDB only supports up to 65536 worth of LBAs). This command can take some time to run (depends on a lot of things I'd rather not get into), and it's atomic.
File overwrites are done similarly -- you need to think about the two conditions that correlate with a file overwrite.
Say you have a file that is 12345 bytes on an NTFS filesystem that uses a 4096-byte cluster (or unit) size. This file takes up 4 clusters/units (4*4096 = 16384). You edit this file, and the new size is 19590 bytes (i.e. is larger). In this case TRIM is not used because there's no removal of data. The file now takes up 5 clusters/units (5*4096 = 20480), despite only 19590 being used.
Understand so far? Okay.
Now alternately let's say you edit the file (in its original state), and the new file size is 7810 bytes. The underlying filesystem shortens the number of clusters/units associated with this file -- it now only requires 2 clusters/units (2*4096 = 8192). The original 2 are no longer used by anything. TRIM is issued on the LBA range that makes up those 2 no-longer-used clusters/units.
TRIM does not operate on byte ranges, it operates on LBA ranges, and the filesystem knows which LBA ranges aren't in use based on file deletions or resizing or whatever else.
So like I said: understanding what TRIM does actually answers the question. 
If you have concerns or questions over how your operating system (Windows) behaves, you should ask Microsoft. The best I can find is a Windows 8 API document that briefly describes the behaviour (see last paragraph of "What is TRIM?") except it doesn't mention file truncation (documentation author or engineer who wrote that did not think about the bigger picture): »msdn.microsoft.com/en-us/library···%29.aspx
--
Making life hard for others since 1977.
I speak for myself and not my employer/affiliates of my employer. |
|
koitsuPremium,MVM
join:2002-07-16
Mountain View, CA
kudos:19 | reply to Hyrules
said by Hyrules:I having this issue with my Corsair SSD. It's the 480 GB. Corsair Force GT 480. with firmware 5.03. It suddenly drop from the controller randomly causing windows 7 to BSOD with the 0xf4 error code. I'm thinking of buying the pro version is the samsung SSD 512 GB.
Your issue is completely, entirely, 100% unrelated to this thread topic. For someone with CompTIA A+ certification this saddens me greatly; I'm sighing and saying "dime a dozen". :/
--
Making life hard for others since 1977.
I speak for myself and not my employer/affiliates of my employer. |
|
