Samsung SSD 840 (250GB) Review
by Kristian Vättö on October 8, 2012 12:14 PM EST- Posted in
- Storage
- SSDs
- Samsung
- TLC
- Samsung SSD 840
Lower Endurance—Why?
Below we have a diagram of a MOSFET (Metal Oxide Semiconductor Field Effect Transistor). When programming a cell, voltage is placed on the control gate, which forms an electric field that allows electrons to tunnel through the silicon oxide barrier to the floating gate. Once the tunneling process is complete, voltage to the control gate is dropped back to 0V and the silicon oxide acts as an insulator. Erasing a cell is done in a similar way but this time the voltage is placed on the silicon substrate (P-well in the picture), which again creates an electric field that allows the electrons to tunnel through the silicon oxide.
While the MOSFET is exactly the same for SLC, MLC and TLC, the difference lies in how the cell is programmed. With SLC, the cell is either programmed or it's not because it can only be "0" or "1". As MLC stores two bits in one cell, its value can either be "00", "01", "10" or "11", which means there are four different voltage states. TLC ups the voltage states to eight as there are eight different combinations of "0" and "1" when grouped in groups of three bits. Below are diagrams showing the graphical version of the voltage states:
SLC
MLC
TLC
The above diagrams show the voltages for brand new NAND—everything looks nice and neat and the only difference is that TLC has more states. However, the tunneling process that happens every time the cell is programmed or erased wears the silicon oxide out. The actual oxide is only about 10nm thick and it gets thinner every time a smaller process node is introduced, which is why endurance gets worse as we move to smaller nodes. When the silicon dioxide wears out, atomic bonds break and some electrons may get trapped inside the oxide during the tunneling process. That builds up negative charge in the silicon oxide, which in turn negates some of the control gate voltage when the cell is programmed.
The wear results in longer erase times because higher voltages need to be applied for longer times before the right voltage is found. Remember, the controller can't adjust to changes in program and erase voltages (well, some can; more on this on the next page) that come from the trapped electrons, cell leakage, and other sources. If the voltage that's supposed to work doesn't, the controller has to basically go on guess basis and simply try different voltages before the right one is found. That takes time and causes even more stress on the silicon oxide.
The difference between SLC, MLC, and TLC is pretty simple: SLC has the fewest voltage states and hence it can tolerate bigger changes in voltages. With TLC, there are eight different states and hence a lot less voltage room to play with. While the exact voltages used are unknown, you basically have to divide the same voltage into eight sections instead of four or two like the graphs above show, which means the voltages don't have room to change as much. The reason why a NAND block has to be retired is that erasing it starts to take too long, which impacts performance (and eventually a NAND block simply becomes nonfunctional, e.g. the voltage states for 010 and 011 begin to overlap).
There is also more and more ECC needed as the NAND wears out because the possibility for errors is greater. With TLC, that's once again a bigger problem because there are three bits to correct instead of one or two. While today's ECC engines are fairly powerful, at some point it will be easier to just retire the block than to keep correcting errors.
Facebook
Twitter
RSS
86 Comments
View All Comments
roedtogsvart - Monday, October 8, 2012 - link
Might have to finally replace my Intel 320 with this. Very nice review KristianOld_Fogie_Late_Bloomer - Monday, October 8, 2012 - link
Both the Intel 320 and the Samsung 840 Pro support encryption...does the non-Pro 840 offer it or not? I'm swinging back towards getting an 840 Pro for my laptop because of the lack of encryption in consumer 830 drives.On a side note, I was disappointed to discover recently that my current desktop motherboard doesn't support hard drive passwords, which is a shame given the fact that my system disk (an Intel 320) does...oh well.
ekon - Monday, October 8, 2012 - link
The ATA password-based encryption these drives use is beset with issues, flaky motherboard support being just one.http://communities.intel.com/thread/20537
On top of which, it's very poorly documented, and notice how reviews give it nothing more than a passing mention without testing the feature. A TrueCrypt/DiskCryptor alternative it is not :-/
Samus - Tuesday, October 9, 2012 - link
I've had my 320 for two years, no problems with encryption, ever.And I second roedtogsvart that I may FINALLY replace my aging SSD with one of these. I've been thinking Intel 520 for awhile but still don't trust Sandforce, even with Intel at the helm.
Samsung and Crucial are keeping the controllers simple, which is why they have the most reliable drives outside of Intel's original X25-M/320 SSD's.
Old_Fogie_Late_Bloomer - Tuesday, October 9, 2012 - link
When you say "these drives", do you mean the Intel 320, the Samsung 840 Pro, or both? My T430 supports hard drive passwords for the expressly-stated purpose of allowing for encrypted SSDs, which is why I'm looking at the 840 Pro in particular.It seems like the only downsides would be having to enter the password to use the computer, and having to use a motherboard that supports HD passwords to access the drive (so I couldn't just plug it into my current desktop if something were to happen to my laptop and it were off for repair or whatever).
ruberbacchus - Saturday, October 20, 2012 - link
On one hand, it seems to me that the potential gains performance-wise with hardware-based rather than software-based encryption are enormous; particularly so on older computers with slower processors. On the other hand, an encryption solution that is not properly documented as well as thoroughly verified and verifiable does simply not exist as a solution; confidence in the implementation is essential to the deployment of encryption. Unfortunately, it seems impossible to obtain clear references to how encryption works in this new class of SSD's. I read somewhere that it should conform to the OPAL standard, which mandates hardware-embedded keys. I am not sure this is such a great idea, for two reasons:- the keys will potentially be known to the manufacturer;
- the keys will be open to physical attacks on the chip controller.
With TrueCrypt e.g., the user himself generates the encryption keys, the master key used for encrypting the data as well as the header key used to wrap the master key. The latter is tied to the user's password. With hardware-based encryption, passwords may be used for authentication, but the keys will not necessarily be derived from the password. Unchangeable encryption keys weakens the security. In order for the system to be trustworthy, the user should be able to generate and re-generate at will their own encryption keys. At the moment it is not clear this is the case.
GreenReaper - Wednesday, August 8, 2018 - link
It's true. Although not necessarily the fault of the drives, either. My leased server was taken offline by the installation of SSDs which the hardware (or rather, its firmware) couldn't handle. I dumped the firmware logs and traced it to commands relating to ATA security. Don't know if it had been set by the previous users of the SSDs or whatever, but it managed to freeze up the entire controller.A5 - Monday, October 8, 2012 - link
I guess I'm old fashioned and don't throw away my laptop every 6 months, but a 3.5 year lifetime seems really really short.crimson117 - Monday, October 8, 2012 - link
Agreed; maybe for a video card, but I expect longer than 3.5 years out of my hard drives.repoman27 - Monday, October 8, 2012 - link
Really? I rarely expect a consumer grade HDD to last much longer than that these days. In fact, I've seen an alarming trend of drives failing just past the 2 year mark, and still within the 3 year warranty period. (RMA'd one last week even.)