Flash memory is built from tiny memory cells that, thanks to vertical stacking technology, can be packed into remarkably dense, compact chips. This topic covers how that vertical NAND architecture works and why it was a turning point in storage capacity.
Vertical NAND Flash Memory
This right here is Flash. It's made up of a lot of little transistors. Actually, they're memory cells. These little memory cells are much smaller than this transistor right here, but I'm going to use this as an example of a memory cell.
Let's say we have a bunch of these and I'm going to stack them side by side. It would actually take up a lot of space to stack these horizontally. Well, what someone did is they figured out, or invented, a way to stack these vertically instead, kind of flip the design of it. And the flipping of the design of this made a huge difference on how much information we could store for every single one of these little chips that we have. It made us able to cram a lot of information into one of these little tiny chips here, and so it was a big step forward.
I will say that the NAND part of this is really not a big part of our conversation. NAND is just a gate. It's a NOT AND gate, and so this is what that's represented of. It's in contrast to a NOR gate. So really this technology has to do with NAND, but that's not the most interesting part of how this works.
Before 2007, our memory cells were stacked side by side, and so it took up a lot of space from a horizontal standpoint. What a vertical NAND is, is we converted it to standing it upright. But that's not the only thing that we did. We redesigned these memory cells to be in columns, and they became very cylindrical. This made us be able to be really efficient with the way we stored our data.
So each one of these would be a memory cell, and they're cylindrical in shape and size, so they're round, and so this was able to store a bunch of data this way. And then we would access each one of these memory cells one at a time. This allowed us to squeeze things down into a much smaller compact area.
I'm on this storagenewsletter.com to show you a diagram that they have, and this diagram shows these cylindrical memory cells that we have right here. Remember, each one of these memory cells, right now we're at triple level cells, can store three bits of information, and then we stack them on top of each other, creating this matrix of data. To access the data on any one of these cells within the matrix, we have to specify a certain X, Y and Z coordinate in order to understand and read that piece of data. And so we go through and we are able to capture different data within this matrix.
I did jump on one other site here, Tom's Hardware, so that we can see this from a little different perspective. Here we have those memory cells all stacked up, those vertical NANDs, and we see it right here too. So there's two stacks of them, right here and here, and then within there, all of this really is so that we can access that matrix. So there's still a lot of components that go in here just so we can select the proper memory cell and read from that memory cell.
It really does quite blow my mind how much memory we can stuff on these little tiny chips, and how much we continue to improve to put more and more on it. It's really quite crazy how far we've come with this technology, and thank goodness too, because it's a huge part of how we operate and how fast our computers have gotten because of this.
TechKnowSurge builds IT and cybersecurity professionals through hands-on, concept-first training built around real understanding — not memorization. Free interactive tools, structured programs, and 25+ years of real-world experience, all in one place.
Explore free tools and programs →