Lower Endurance: Hardly an Issue

With perfect wear-leveling and write amplification of 1x, you would get 256,000GiB (that's ~275TB) of writes out of a 250GB Samsung 840 with TLC NAND and 1,000 P/E cycles. That is still a lot but wear-leveling and write amplification are never perfect. Giving any specific numbers for endurance is hard because every drive behaves differently and users have different workloads, but it's unlikely for a light consumer workload to see more than 10GiB of writes per day. That's 3,650GiB per year, which is only 1.4% out of 256,000GiB. In the real world NAND writes will be bigger than host writes but even with a write amplification factor of 10x, you will only end up writing 36,500GiB each year and exhausting ~143 P/E cycles out of the available 1,000. In other words, it would take roughly seven years for you to wear out the NAND.

SSD Lifetime Estimation
NAND MLC—3K P/E Cycles TLC—1K P/E Cycles
NAND Capacity 128GiB 256GiB 128GiB 256GiB
Writes per Day 10GiB 10GiB 10GiB 10GiB
Write Amplification 10x 10x 10x 10x
Total Estimated Lifespan 10.5 years 21.0 years 3.5 years 7.0 years

For the 120GB Samsung 840, the lifespan is half of the 250GB model but we are still talking about years. Samsung doesn't offer a 60/64GB Samsung 840, although that makes sense as it wouldn't be hard to wear that out in less than three years, which is the warranty Samsung gives to the 840 SSD.

DSP to the Rescue

However, there is actually more to SSD endurance than just P/E cycles and write amplification. There has been a lot of talk lately about digital signal processing (DSP) in the industry, which is supposedly the solution for lower endurance NAND.

The basic idea behind DSP is very simple: you read changes in voltages and adapt to the changes. As I mentioned in the previous page, the voltages change as the NAND wears out and if your controller can't adapt to the changes, you'll be stressing the NAND even more. Each time you're trying to program or erase the cell, you are wearing it out, so you don't even have to succesfully program or erase the cell to cause damage. That's why the guess and test process for writing to NAND is so harmful; it may take multiple tries and each try will wear out the NAND even more.

 

Graphical presentation of a change in voltage state

However, if your controller can read the changes in program and erase voltages, you will know what voltages to use to program/erase the cell. Even though DSP doesn't make NAND immortal, it causes a lot less stress on the NAND, allowing it to last for more P/E cycles than what you would get without DSP.

Again, it's hard to give out any specific numbers of DSP usefulness in real world, but for example STEC is claiming that their CellCare technology can extend the endurance of regular 3K P/E cycle MLC up to as much as 60K. I've heard unofficial figures as high as 100K for some companies' DSPs, but I would take all figures with a grain of salt until they are tested by a third party. Either way, even if a good DSP is only able to double the endurance of NAND, it's a huge deal as we move to even smaller process nodes and possibly even more bits per cell.

Lower Endurance - Why? The Samsung SSD 840
POST A COMMENT

84 Comments

View All Comments

  • roedtogsvart - Monday, October 08, 2012 - link

    Might have to finally replace my Intel 320 with this. Very nice review Kristian Reply
  • Old_Fogie_Late_Bloomer - Monday, October 08, 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.
    Reply
  • ekon - Monday, October 08, 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 :-/
    Reply
  • Samus - Tuesday, October 09, 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.
    Reply
  • Old_Fogie_Late_Bloomer - Tuesday, October 09, 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).
    Reply
  • 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.
    Reply
  • A5 - Monday, October 08, 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. Reply
  • crimson117 - Monday, October 08, 2012 - link

    Agreed; maybe for a video card, but I expect longer than 3.5 years out of my hard drives. Reply
  • repoman27 - Monday, October 08, 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.) Reply
  • travbrad - Wednesday, October 10, 2012 - link

    I had a 80GB WD that lasted 8 years without failing. I eventually had to stop using it simply because it was too slow. I also had a 250GB WD drive that I used for 5 years (then switched to all SATA). Now I have a 640GB drive that I've been using for almost 4 years. My brother has a couple 500GB drives in his system that have been running for 4-5 years as well.

    Maybe I've just been really lucky, but the only drive I've personally had fail in the last decade was a Hitachi drive (obviously selected for cost not quality) in my HP laptop.

    Now at work it's a different story. Those pre-built machines cut every corner they can to bring costs down so they end up with low quality components (especially PSUs). Even in that situation there is a fairly low number of hard drive failures though (considering how old most of the machines are)
    Reply

Log in

Don't have an account? Sign up now