https://blog.korelogic.com/blog/2015/03/24
Some excerpts:
If so, what kind of imaging software should I use?
Some excerpts:
If so, what kind of imaging software should I use?
SSDs have a shelf life. They need consistent access to a power source in order for them to not lose data over time. There are a number of factors that influence the non-powered retention period that an SSD has before potential data loss. These factors include amount of use the drive has already experienced, the temperature of the storage environment, and the materials that comprise the memory chips in the drive.
For client application SSDs, the powered-off retention period standard is one year while enterprise application SSDs have a powered-off retention period of three months. These retention periods can vary greatly depending on the temperature of the storage area that houses SSDs.
Heh, that puts a new twist into inventory management, maybe they should have expiration dates for laptops. "Product must be bought and used before DD/MM/YY."Assume it isn't as big a deal since laptops sit in inventory for a while.
Essentially true. IRL they will meet or exceed those ratings.
Assume it isn't as big a deal since laptops sit in inventory for a while.
Do you often leave disks sitting in a box?
It's potentially true. Problem is there's been so little research in that area (most "lifespan" tests are on write endurance, not potential charge dissipation / data retention). As BonzaiDuck said, it's the data not the SSD that may have problems. Basically the potential problem is : As the cell size shrinks, so does the floating gate size which hold fewer and fewer electrons every time the process node gets smaller. IIRC 15nm NAND stores less than 20 electrons per NAND cell. That's already not very many even with MLC, but with TLC, that works out to just three electrons per voltage state. Not much room for error. Cells also get closer as die sizes shrink and given the key factor in NAND is the capacitance between the control and floating gate, as the cells move closer to each other through die shrinks, the neighboring cells will also introduce capacitive coupling, ie, may interfere more. IIRC, at 90nm, the interference was only around 8% but at 20nm it's up to 40%. At sub 20nm, it's probably higher.
The Samsung 840 slowdown issue is about the same thing - Old files that are rarely accessed or moved around for longer periods seem to lose their status on TLC + 19nm technology. From what I understand the massive slowdowns are due to needing error correction applied in order to retrieve the data which has become marginally readable due to 19nm (few electrons per cell) having to store more voltage states (TLC) = vastly fewer electrons per cell per voltage state compared to MLC on the same process / TLC drives on a larger process.
Personally, the whole Samsung debacle has convinced me of three things : 1. Avoid small process Samsung TLC drives like the plague no matter how cheap they are. 2. 40nm 3D NAND may have longer unpowered data retention times than 16-20nm. 3. Mechanical HDD's are still the best choice for long term offline backups (external backup drives only occasionally plugged in).
Edit: Cerb is right that data retention is at rated max P/E cycles and that it should be higher (in theory). Problem is, I've seen a couple of flash drives (similar technology) be only part readable after 18-24 months unpowered despite the manufacturer's claim "10 year lifespan" (unpowered data retention). As with "100 year lifespan" optical media - the facts do not always meet the manufacturers "accelerated testing" based claims.
I don't know about software that detects the "quality" of the data on a hardware level, but whenever I back anything up / move data between drives I always use TeraCopy (built in source & readback target CRC comparison). It won't help with data degradation (unless you've kept the original data to compare with), but I find it does eliminate potential "silent errors" during the backup phase.What kind of software should I use to check the data integrity of the SSD?
Bottom of the barrel TLC, with questionable controllers. The margins on SD, and small USB sticks (IE, single package for flash and controller), are so low that they can actually get away with selling them cheaper than stringently-binned flash. The controllers are supposed to handle errors as they come across them, so only the most basic functional testing has to be done. Crazy stuff. If it's not specifically high end (like Sandisk Extreme, or Kingston series with MIL spec testing), I don't trust any basic flash devices, nowadays.Problem is, I've seen a couple of flash drives (similar technology) be only part readable after 18-24 months unpowered despite the manufacturer's claim "10 year lifespan" (unpowered data retention). As with "100 year lifespan" optical media - the facts do not always meet the manufacturers "accelerated testing" based claims.
I have a gaming PC that I leave when I'm not in the country (half a year1 year). I recently upgraded to a SSD (love it).
What kind of software should I use to check the data integrity of the SSD?
All in all, there is absolutely zero reason to worry about SSD data retention in typical client environment. Remember that the figures presented here are for a drive that has already passed its endurance rating, so for new drives the data retention is considerably higher, typically over ten years for MLC NAND based SSDs. If you buy a drive today and stash it away, the drive itself will become totally obsolete quicker than it will lose its data. Besides, given the cost of SSDs, it's not cost efficient to use them for cold storage anyway, so if you're looking to archive data I would recommend going with hard drives for cost reasons alone.
i've got 8 ssds on a shelf.
you're telling me they are going to lose data just sitting there?
sounds like fud to me.
In any event, the "data" is just game installs anyway so I really could care less about losing it. Hmm is the correct expression couldn't care less or could care less while I'm asking questions?
That point is defined at the point where the bit error rate is high enough, along with the resting degradation, for the data to last a year (or more). Prior to that point, the bit error rate is lower, and generally follows a steep curve. That error rate directly correlates to how well the flash holds charge, and the amount of correct bits that get written in the first place. Even at the same rate of degradation of charge, it would take much longer to be unreadable, with a miniscule error rate on the write itself.According to Anandtech, " Finally, the drive must retain data without power for a set amount of time to meet the JEDEC spec. Note that all these must be conditions must be met when the maximum number of data has been written i.e. if a drive is rated at 100TB, it must meet these specs after 100TB of writes."
Why is everyone assuming that because drives must meet these requirements at max P/E life, they automatically mean drastically greater data retention times before that point?
Most notebooks, for example, will have the SSD running above 30C, while I can't think of any place I would store any electronics that wouldn't remain under 30C for most of the year, if not all of it. Computer parts shouldn't be stored outdoors, especially in hot climates. A hot climate could mean higher operating temps, but for offline/idle temps, it shouldn't have an effect. IE, being in a hot climate should mean cooling costs more, not that anyone that cares about their stuff should be leaving it in the heat.Also, isn't is more realistic for someone to use a computer in a cooler environment [aircon] and leave it in a hotter place when not in use than the reverse? Especially in hot climates.
Remind me about the lifetimes for data retention in optical discs when they were introduced and for many years afterward.That point is defined at the point where the bit error rate is high enough, along with the resting degradation, for the data to last a year (or more). Prior to that point, the bit error rate is lower, and generally follows a steep curve. That error rate directly correlates to how well the flash holds charge, and the amount of correct bits that get written in the first place. Even at the same rate of degradation of charge, it would take much longer to be unreadable, with a miniscule error rate on the write itself.
Most notebooks, for example, will have the SSD running above 30C, while I can't think of any place I would store any electronics that wouldn't remain under 30C for most of the year, if not all of it. Computer parts shouldn't be stored outdoors, especially in hot climates. A hot climate could mean higher operating temps, but for offline/idle temps, it shouldn't have an effect. IE, being in a hot climate should mean cooling costs more, not that anyone that cares about their stuff should be leaving it in the heat.
OK. I still have yet to see a Verbatim Datalife Plus, or TY, fail, unless horribly abused. Pretty soon, a few that I have around just to see about that will be 20 years old.Remind me about the lifetimes for data retention in optical discs when they were introduced and for many years afterward.
I don't. I assume that if you care about your parts, and/or care about your data, you will attempt to treat them well. If you don't, then you are not going to have good results with any commodity storage technology. If you live somewhere without an infrastructure to allow you that, then you need to be using cloud backup services for any valuable data.Computer prices falling have allowed people in many countries to use them so why should anyone assume that their own operating conditions and usage are typical for everyone?