Samsung SSD 850 Pro (128GB, 256GB & 1TB) Review: Enter the 3D Era
by Kristian Vättö on July 1, 2014 10:00 AM EST3D NAND: How It Works
To understand how 3D NAND helps to keep Moore's Law in action, we first need to go in-depth with the structure. Before we begin, there are a couple of disclaimers I want to get out. First of all, every manufacturer has a different 3D NAND structure, so to avoid information overflow and confusion, I will only talk about Samsung's structure for now. When other manufacturers are ready with their 3D NAND products, I will provide a similar analysis of their structure. Secondly, there is not much detailed information about Samsung's 3D NAND, or V-NAND as they call it, so I am mostly basing my analysis on The Memory Guy's blog post (also known as Jim Handy, a respected semiconductor analyst).
What you are seeing above is a 5-layer 3D NAND string based on Samsung's TCAT (Terabit Cell Array Transistor) structure. It consists of a total of ten cells and similar to 2D NAND each cell is capable of holding one, two or three bits of data depending on whether the NAND is SLC, MLC or TLC. Samsung's current (i.e. second) generation V-NAND has 32 layers, meaning that it is simply a taller tower but to keep the graph easily readable I decreased the number of layers to five.
There is one fundamental difference between Samsung's V-NAND. With 2D NAND, the charge was stored in a conductive floating gate but as you can see in the graph above, there is not one in Samsung's V-NAND. Instead Samsung uses a design called Charge Trap Flash (CTF), which means that the charge is stored in an insulator, which is silicon nitride in this case.
Some of you might remember this rather hilarious slide we used in our V-NAND announcement article and I now have an explanation as to why Samsung used cheese in in (yes, it indeed is cheese). A traditional floating gate works pretty much like a bucket of water. As long as there is not a hole in the bucket, the water stays there without an issue. However, if you get even a tiny hole in the bucket, all the water will sooner than later escape through that hole. It is the same with a floating gate because if there is a hole in the insulating material around it (Inter Poly Dielectric and tunnel oxide), all the electrons in the floating gate will escape through that as the floating gate is conductive and the electrons can move freely. As I mentioned on the previous page, this is exactly what happens when NAND is cycled and wears out because the stress caused by the programming operation causes the insulators to lose their insulating characteristics.
With Charge Trap Flash that does not happen because the electrons reside in an insulator. Samsung compared the bucket of water idea to cheese, meaning that if there is a hole the cheese will not just pour out from it like water would.
As a result, Charge Trap Flash increases endurance as it is not as vulnerable to wear out. I am also told that CTF does not require as high programming voltages as floating gates do (up to 20V), which reduces the stess on the insulators. I wonder if this is simply because the insulators do not have to be that thick anymore (a miminum of 7nm for tunnel oxide and 12nm for ONO) because the charge trap is not as dependent on the surrounding insulators as floating gate is. The reason why such a high voltage is needed for 2D NAND programming is the thickness of the tunnel oxide because otherwise the electrons cannot tunnel through (remember, the electrons are shot through an insulator). On the other hand, the tunnel oxide cannot be made any thinner than 7nm or otherwise data retention takes a massive hit (NAND is really an engineers dream, isn't it? Touch one thing and you break another).
Another interesting tidbit regarding Samsung’s V-NAND is the usage of a high-K dielectrics. IMFT did a similar switch when they moved to 20nm node but all the other manufacturers, as far as I know, have stayed with an inter poly dielectric (IPD) design. I went into detail about high-K dielectrics in our Crucial MX100 review but I will provide a quick summary here. Basically, a high-K dielectric is a material with a high dielectric constant, which is quite surprisingly known as K. The benefit of a high-K dielectric over a normal dielectric (like oxide-nitride-oxide i.e. ONO in NAND) is that it provides a higher capacitance between the word and bitlines. Similar to 2D NAND, the capacitance between the control gate and the floating gate, or charge trap in the case of V-NAND, is still the key factor for operation. The usage of high-K dielectrics gives Samsung a bit more headroom in terms of vertical scaling as layers can be stacked closer to each other and it also improves endurance as high-K dielectrics have less leakage compared to normal dielectrics.
Otherwise the basics of V-NAND are very similar to 2D NAND. To program a cell, the bitline is grounded (i.e. held at 0V) while a high voltage is placed on the wordline. That makes the electrons that are floating in the bitline to tunnel through the silicon dioxide to the charge trap a.k.a. silicon nitride.
I made a very simple graph of V-NAND programming, which should help you to understand the process. The elements are the same as in the tower graph with the difference that it is just one layer and I cut the tower in half to fit it on the page. The purple balls are electrons.
Reading from V-NAND works exactly the same as with 2D NAND. The wordline of the cell-to-be-read is held at 0V while different voltages are applied on the bitline. Once the correct voltage is found, the cell will conduct and the sense amp will read the data depending on what the voltage was (the voltage determines the voltage state, i.e. what the value is).
Facebook
Twitter
RSS
160 Comments
View All Comments
GTVic - Tuesday, July 1, 2014 - link
Wondering if the 3D V-NAND has an effect on heat produced by each chip?Laststop311 - Wednesday, July 2, 2014 - link
If only this drive would of been a pci-e 3.0 x4 interface with nvme. We would of finally had a worthy upgrade. Yes it's performance is better than the 840 evo but you can get the 840 evo 1tb for 400 dollars even less if you catch it on sale. So you can get 2x 1tb 840 evo for almost the same price as 1 850 pro. If you compare an 840 evo with 25% OP to a regular 7% on the 850 the 840 evo looks just as good so 2x 825GB drives with 25% OP on each drive costs you 750-800 depending on the deal you get vs 700 for 1000GB. I would rather pay 800 for 1650GB than 700 for 1000GB with performance being nearly identical. I get an extra 650GB (which at 50 cents per GB is another 325 dollars worth of ssd) and tons of over provisioning to give the drive equal or better performance for only 100 dollars more (possibly less as the 840 evo has often sales)Sorry Samsung but 2x 1tb 840 evos with 25% provisioning gives me better or equal performance and a whole 650GB of extra storage for only 100 dollars more. At 50 cents per GB you get 325 dollars worth of more storage capacity for only 100 dollars more and thats with the 25% over provisioning which basically negates the performance increase of the new drives.
The only way samsung could of made this drive worth that money is if they had the drive on a pci-e 3.0 x4 interface with nvme instructions. I'm sure there will be tons of idiots who just buy it cause it's the latest drive. But if you use your brain you can see the 840 evo is still the best SATA drive when the cost/performance ratio is taken into account. 2x 1tb evo's in 25% OP mode gives you 1650GB and costs u 100 dollars extra or less and gives you the same performance or even better compared to 7% 850 pro 1000GB especially when raid 0 is taken into account. I'll take 1650GB over 1000GB if it's only 100 more and performance is equal or better easy choice.
Laststop311 - Wednesday, July 2, 2014 - link
Since samsung is stacking vertically now what they should of done is made a super huge 2tb drive to differentiate themselves from all these other drives. A 2tb drive for 1400 is a little more acceptable than 1tb for 700 simply because it's the only single drive with 2TB capacity.I can see great things happening with vertical stacked nand tho. When this process matures we should be seeing nand drives surpass spinning hard drives in capacity. When samsung has those 1tbit dies its planning for 2017 we should be able to have 8-12TB SSD's
althaz - Wednesday, July 2, 2014 - link
It's a new product and it's priced highly. Eager early adopters who want to move on to the latest and greatest will buy now, value-concious people will buy the 840s. Inventory of the 840s will get eaten up and the 850s will drop in price.This is what happens whenever any product is released, basically ever.
asmian - Wednesday, July 2, 2014 - link
Sorry, but neither this nor the EVO will get my money. Performance is all very well, even if only a tiny handful of users with extremely niche workloads will ever notice any difference, but all this extra reliability at a price premium means NOTHING without simple power loss protection. Restricting something so basic to "Enterprise" class products is the real gouging here by Samsung, not the price.If Crucial can provide that protection on the CHEAPEST drives in their class (M500/M550 and IIRC MX100 too) with performance that is not gimped as a balance, then there is no excuse for Samsung not to. This should be a no-compromise baseline for all SSDs going forward, and Anandtech should push hard for that - users should as well, by voting with their wallets and refusing to buy drives, however fast and powerful, that do not provide power loss protection as a basic feature.
bsd228 - Wednesday, July 2, 2014 - link
Though I agree it is a highly useful feature, it is far less significant to those of us using a good UPS. So I can't agree that it's a no compromise feature.romrunning - Wednesday, July 2, 2014 - link
What you aren't taking into account is the fact that the 850 Pro has MUCH higher endurance, and it's also more consistent. Those two items bring it more into the Intel DC 3500/3700 series type of class. It's not just a sheer performance comparison....and if you thing the 850 Pro is expensive, price out the larger Intel DC S3700 drives.
FunBunny2 - Wednesday, July 2, 2014 - link
"Real" Enterprise SSDs don't even have an MSRP. You negotiate with the vendor, and hope for the best. Now that Texas Memory is in IBM, and Fusion-io in SanDisk, with Violin likely to go somewhere. The conundrum is V.NAND's impact on flattening the curve between Enterprise and Commodity/Retail. At one time, a mainframe had bespoke 14" behemoth hard drive subsystems (in the case of IBM, run by the equivalent of a PDP-x). In due time, binned commodity 3.5" drives are now used.Samsung could well be the driving force to regularize solid state storage. The remaining issue is whether the file system hand waving will be dumped in favor of direct NVM persistence? Samsung, or whoever, likely couldn't care less.
romrunning - Wednesday, July 2, 2014 - link
It's been enjoyable to see a lot of "new" flash memory storage vendors pop-up. More competition is always good in that enterprise space.I've been looking forward to having more SSDs options available to servers at much better pricing. Solid storage advances have a trickle-down effect. If I can put an array of these Samsung 850 Pros into a server and achieve near "enterprise" performance, then that forces Dell/HP/etc. to drop their own SSD pricing.
watersb - Wednesday, July 2, 2014 - link
Fantastic detail of 3D NAND design and why it matters. Thanks very much!