What do you do if you need very fast data processing and don’t want to…
By Chris Bross, Chief Technology Officer
Anyone dealing with solid-state drive (SSD) technology should know a few things about how these unique devices work and what they are supposed to do, so we’ve prepared the following “cheat sheet” to help you understand the technology behind these advanced data storage machines.
Unlike a traditional electro-mechanical drive, a solid-state drive has no moving parts. The SSD uses NAND flash memory chips to store data. And, instead of writing data to spinning platters (HDD or hard disk drive), an SSD stores data to an array of those memory chips.
Early SSDs had capacities much smaller than the average HDD, and at a much higher cost. Today, there are enterprise level SSDs that can store up to 16TB of data on one device and the cost differences are falling.
The use of 3D NAND chips—like the 3D Xpoint from Intel/Micron, V-NAND from Samsung and 3D NAND from Toshiba/Sandisk—is on the rise as manufacturers try to squeeze more information into smaller and smaller spaces.
With 3D NAND technologies, the memory cells are stacked vertically in layers to achieve higher densities—a move that also helps reduce manufacturing costs.
At the enterprise level, manufacturers like PURE Storage and SolidFire/NetApp are making RAID systems that link multiple large SSDs together to provide even greater storage capacity.
Here is a short list of key technical terms regarding solid-state drives:
The process through which the controller balances the writing and erasing of data across the storage area of an SSD. The idea—similar to rotating the tires on your car and including the spare tire in the rotation—is to spread the wear evenly on all of the tires, but with the SSD the usage is spread across all of the available cells to prolong the operating life of the media.
An action that identifies which blocks of data are no longer needed on a solid-state drive and can be wiped for re-use. This process happens quickly and, in most cases, automatically.
A memory management process through which unused portions of a solid-state drive are identified, modified and then utilized for storage of new information.
The smallest unit of measurement used for computer data. It is a single binary value, either 0 or 1. Bits are most commonly used in reference to data transfer rates (the speed in which data is moved from one place to another).
Single-Level Cell, Multi-Level Cell, Triple-Level Cell
The difference in these terms is somewhat self-explanatory. With a single-level cell (SLC), only one piece of information (one bit) is contained within a storage cell. Multi-Level cells (MLC) can hold two pieces (bits) of information and TLCs can hold 3 bits of information per cell.
A controller is a device used in both HDDs and SSDs to communicate with the host device and manage the data storage process. An SSD controller also manages more of the technical operations, including wear leveling and garbage collection (see definitions above).
P/E or Program/Erase Cycle
A way to measure how many times the storage areas of a solid-state drive can be written to, deleted and then re-used.
A logical controller interface through which the solid-state device communicates to a computer via physical connection from the SSD’s M.2 or U.2 to the PCI Express slot on the motherboard of that computer.
NVM Express is similar to the USB connector on most portable hard drives. It was designed to maximize the bandwidth and performance of solid-state storage devices. This interface allows levels of parallelism found in modern SSDs to be fully utilized by the host hardware and software. As a result, NVM Express reduces I/O overhead and brings various performance improvements in comparison to previous logical device interfaces, including multiple, long command queues and reduced latency.
This physical interface, formerly known as the Next Generation Form Factor (NGFF), is a specification for internally mounted computer expansion cards and associated connectors. It replaces the mSATA standard, which uses the PCI Express Mini Card physical card layout and connectors.
U.2 is the upgraded external physical cable interface, formerly known as Sata Express, for connecting external storage devices to the PCI Express port on the backplane of the computer.
PCI Express (PCIe)
PCI Express is a high-speed serial connection on a computer that acts more like a network than a traditional bus. A bus is a communication system that transfers data between components inside a computer, or between different computers. Instead of one bus that handles data from multiple sources, PCI Express (PCIe) uses a switch that controls several point-to-point serial connections. These connections lead to the devices where the data is needed. Every device has its own connection, so there is no sharing of bandwidth like that which occurs on a normal bus. PCle has other advances over the older standards as well, including higher throughput, smaller physical size, more detailed error detecting and reporting mechanisms along with greater support for I/O virtualization.