How do you define SSD? SSDs (referred to as solid-state drives or solid-state disks) are storage devices that store persistent data on solid-state flash memory. SSDs have no moving parts involved. Rather, an SSD is made up of semiconductor memory organized as a disk drive, using integrated circuits (ICs) rather than magnetic or optical media.
But, if you are not familiar with SSD storage, you may come across a lot of terms that may be confusing at first glance.
We’ve compiled this list of important solid-state storage terms. The list includes a few terms you should know already as well as some that may seem obscure, but are no less essential to the world of SSD.
1. What is SSD overprovisioning?
Just as it sounds, this is the inclusion of extra storage capacity in flash solid-state drive. That extra capacity is not visible to the host as available storage. SSD overprovisioning has one major benefit: It can increase the endurance of a solid-state drive by distributing the total number of writes and erases across a larger population of NAND blocks and pages over time.
2. What is PCIe solid-state storage?
PCIe is a high-speed expansion card format that is installed directly in a server. PCIe-based solid-state storage typically performs better than server-based SATA, SAS or Fibre Channel solid-state drives because of the direct connections. It’s an ideal choice for applications with intensive I/O requirements such as online transaction processing and data warehousing.
3. What is Tier 0?
Tier 0 is the fastest and most expensive level of storage in the storage hierarchy. It’s ideal for an enterprise that requires select applications to be quickly accessible. The storage hierarchy changed with the addition of tier 0 storage. It marked a change from simply moving less active data to slower, less expensive storage to focusing on efforts to move more active data to faster, more expensive storage and SSDs are driving that trend.
4. What is resistive RAM (RRAM)?
Resistive random access memory is a form of nonvolatile storage that operates by changing the resistance of a specially formulated solid dielectric material. RRAM devices contain a component called memristor, which is a contraction of “memory resistor” and its resistance varies when different voltages are imposed upon it. One of the main advantages of RRAM is its high switching speed compared to other nonvolatile storage technologies. One challenge that RRAM devices presents to engineers who want large-scale development of memristor technology is the occasional formation of unintended filaments (otherwise known as “sneak paths”).
5. What are IOPS?
Input/output per second (IOPS) is the standard unit of measurement for the maximum number of reads and writes to non-contiguous storage locations.
6. What is solid-state storage program-erase cycle?
The solid-state storage program-erase cycle is a sequence of events where data is written to solid-state NAND flash memory, then erased, and then rewritten. Program-erase cycles serve as a means for quantifying the endurance of a flash storage device. Flash memory devices have a limited number of PE cycles because each cycle causes some physical damage to the medium. The damage accumulates, eventually leaving the device useless. The number of PE cycles a particular device can sustain before problems occur is varied with the type of technology.
7. What is TRIM?
TRIM is a SATA interface command that tells the NAND flash solid-state storage device which data to erase. The TRIM command allows the OS to notify the SSD which data in a set of pages can be overwritten, allowing the SSD controller to manage the erase process between the time when the host initiates a delete and the next write. By removing erase from the write process, writes occur faster.
8. What is a hybrid hard disk drive (HDD)?
The hybrid hard disk drive is a spinning hard disk drive with onboard NAND flash memory. The NAND flash serves as a non-volatile cache, allowing faster access to data.
9. What is multi-level cell (MLC)?
MLC is flash memory that stores more than one bit per cell. It’s less expensive than single-level cell (SLC) flash, which makes it a desirable option for consumer-grade solid-state storage. It does have a higher bit error rate than SLC flash because there are more opportunities for misinterpreting the cell’s state.
10. What is a flash controller?
Flash controllers are the part of flash memory that communicate with the host device and manage the flash file directory. The controller is responsible for wear leveling, error correction and garbage collection.
This was first published in May 2012