Cloud Storage vs. Online Backup: Which is Best?
August 11, 2015
How to Recover RAID 5 Data
RAID-5 Setup Tips: How to Avoid the Data Recovery Lab
August 17, 2015

Video: What’s in a Solid State Drive?

We get plenty of data recovery cases for solid state drives every year. They are fascinating and fairly complex pieces of technology. With price points dropping and technology improving every year, more and more people are buying SSDs. In fact, plenty of people buy solid state drives simply so they can have faster boot times by putting their operating system on one (actually, I did that).

 

With growing consumer interest in solid state drives, we thought it would be beneficial to make a video explaining the various components inside a solid state drive. It can seem overwhelming at first, but the truth is, there are really only five or six components to touch on.

 

Without further ado, here’s our video on what’s inside a solid state drive.

 

 

Full transcript:

Hi, I’m Jesse Moryn with Gillware Data Recovery. In recent years, we’ve seen some great advances in solid state technology- leading to better, more affordable solid state drives. A lot of consumers have been taking advantage of this by putting solid states in their computers. However, they don’t really know much about them. So today, we’re going to take apart this Intel 320 Series SSD and take a look at just what’s inside.

 

So here we have a closer look at our drive. You can see the name and the size- 160 Gigabytes. We’re going to go ahead and take this drive apart through the miracle of a bad jump cut. There we go, we’re good to go. Now you can see the inside of the drive, uh- the NAND flash chips here, the processor or controller right here, and the SATA over here. Those are the big three that you’re going to find in pretty much every solid state. There’s a few other components we can talk about that are not necessarily present in, uh- some solid state drives, but we can talk about them anyway because they’re interesting.

 

So NAND flash memory is the first component we’ll discuss- and it’s the storage for our drive, also known as non-volatile memory, since it doesn’t need a power source to keep the data. This can be compared to the spinning platters or disks of a hard drive, but instead of storing the information on a magnetic substrate, it’s stored in the stationary NAND flash memory chips.

 

The lack of any moving parts is actually why we call them solid state drives, they’re in a solid state, there’s no need for spinning disks or rapidly moving read/write heads. The reason this memory in particular is called NAND flash memory, is because the memory uses Not-And logic gates, or NAND for short. Now, the way NAND flash memory stores data is through the use of floating gate transistors.

 

Floating gates are electrically isolated, meaning they have no electrical contacts because they’re insulated by oxide layers. Because of this, electrons are able to stay within the floating gates for years. There are multiple ways we can get electrons into the floating gates, but the basis is by utilizing quantum tunneling to get them through the oxide layer from the channel.

 

Because those electrons aren’t leaving the floating gate anytime soon, we use the presence or absence of those electrons to represent the 0’s and 1’s of our data. This is called SLC or single-level cell NAND since there’s only 2 possible charge states. We can also use different amounts of electrons to represent different charges and thus increase our data storage density.

 

This is how MLC or multi-level cell NAND works. All consumer grade drives use MLC NAND, including this one. Of course the full explanation on solid state storage is much longer than that, but that’s the basics of it.

 

Next up we have our processor, also known as the controller. This little guy is the fastest and debatably most important part of the drive since it’s responsible for interfacing between the other components, and is essentially the brains for the drive. When I say it’s the fastest part of the drive, I mean it executes operations on the order of nanoseconds, a nanosecond being one billionth of a second.

 

It performs operations that are written into the firmware such as all the reading, writing, erasing, encryption, garbage-collection, wear-leveling, and over-provisioning. Wow. I apologize if you don’t know what some of those operations are, but just know they’re important. Without the controller to perform all these functions, your solid state drive would pretty much just be an elaborate hunk of scrap metal. So yeah, it’s pretty important.

 

Third, we’ll look at the SATA interface which is this part here on the end with the gold pin connectors. This is the part that connects up with the computer and is connected through SATA power and data cables. Power is the big one, data is the small one. There’s not much to explain about the SATA interface other than it’s the bridge that allows the transfer of data between your SSD and your computer, as well as brings power to the SSD.

 

Finally we’ve gotten to the components that aren’t in some solid state drives, but can certainly be useful for the ones that do have them. That’s not to say that all SSD’s even need these components, since some of them work slightly differently. But anyway, these components are the SDRAM and capacitors.

 

The SDRAM, or synchronous dynamic random access memory, is the volatile memory for your SSD. I mentioned earlier that NAND flash memory is non-volatile, meaning it doesn’t need a power source to store data. Well you’ve probably figured out that because it’s volatile memory, SDRAM does require a power source to store data. If there’s no power going to your SSD, like when your computer is turned off, the SDRAM flushes the data out and forgets everything.

 

So what’s the point of having it if it’s seemingly less functional than NAND? Well, it isn’t less functional than NAND, it’s actually faster and allows the controller to run programs quickly. The SDRAM only serves as working memory for your controller, and it only puts data here that’s essential for whatever the task at hand requires.

 

You can imagine that volatile memory requiring power to store data can cause some problems, and you’re right.
If there’s a sudden loss of power, whatever is stored in the RAM can be lost forever, which is similar to the reason you’re not supposed to just pull a USB out of your computer before ejecting it.

 

This can corrupt files and cause data loss if the drive had data in the RAM when you cut the power. This problem leads me to the final components which are the capacitors. In the event of a sudden loss of power, these capacitors serve as an emergency power source to flush any data in the RAM back into the NAND flash memory so you don’t lose it forever.

 

They’re a cool failsafe to prevent data loss. Of course one component I didn’t mention is the PCB, or printed circuit board, and that’s this green thing underneath all the other pieces. This is the component that allows all the other parts to work together and communicate with each other so your drive works exactly how it’s supposed to.

 

That concludes our video on what’s inside a solid state drive. Here at Gillware, we want to thank you for watching and hope you enjoyed it, but most importantly, we hope you learned something.

 




ssd-form-factors-buyers-guide-cta

//

10 Comments

  1. […] solid state drives, and most other flash memory devices. For more information on NAND flash, watch our video on SSDs. The section on NAND runs from 1:00 to […]

  2. Al Parker says:

    I did not watch the video. I learned much about SSDs that I did not know. Thanks.

  3. […] causing HDD prices to double globally. This price increase also caused many consumers to purchase solid-state drives (SSDs) in lieu of […]

  4. […] causing HDD prices to double globally. This price increase also caused many consumers to purchase solid-state drives (SSDs) in lieu of […]

  5. […] couple years down the line, it is clear that many enterprise data solutions will require the use of solid-state drives to accommodate the vast quantities of […]

  6. […] couple years down the line, it is clear that many enterprise data solutions will require the use of solid-state drives to accommodate the vast quantities of […]

  7. […] you want to know more about how NAND flash works, we recommend checking out our video “What’s In a Solid State Drive?” It’s slightly too complicated to briefly explain right here, but just know that your […]

  8. Thor says:

    Very informative. A few quick questions. The capacitors are meant to hold enough energy so that data may be flushed back to the drive in case of power failure. Are they capable of flushing the entire capacity of the volatile RAM? What are the limits and how reliable is this. To put it succintly, if the power is just cut off, does this allow for any and all data not on the solid state to be written. If so, then why dont all devices carry capacitors?

    • Hi Thor,

      The level of built-in power-loss protection varies from drive to drive and is designed to be just enough for the SSD to settle its affairs, so to speak, when power is suddenly lost. For example, the capacitors in consumer-grade Micron SSDs provide enough power for an additional 1 ms of device operation. The design team at Micron must have concluded that this would be sufficient in most cases based on the anticipated usage pattern of the typical consumer.

      Enterprise-level drives include expanded power-loss protection, which is one of the reasons we always recommend them to clients seeking the most-reliable storage. Appallingly, there are SSDs on the market that feature no power-loss protection at all. Corner-cutting like this is common in cheap SSDs which trade reliability for a lower price.

Leave a Reply

Your email address will not be published. Required fields are marked *