How Are Things Stored on a Flash Drive?

Flash drives are a compact form of storage that stores information differently from other types of storage media.

As technological capabilities increase and become more accessible to the general public, there is a growing need for larger storage capacities. To make this possible, a large amount of data needs to fit into a small device. Flash drives are a solution to this problem, offering the ability to fit entire media collections onto a device the size of a thumb. To accomplish this, special data storage techniques are required.

Background

Data storage comes in a variety of formats, each adapted to the particular task to which it will be applied. One example of this is Random Access Memory (RAM), which is used to store applications that are currently running on a computer. RAM is suited to this task due to its high speed, but is unsuitable for long-term storage because it is cleared when it is not plugged in to a power source. Flash memory is a subtype of a kind of memory called EEPROM, or Electrically Erasable Programmable Read-Only Memory.

Attributes

Flash memory has a number of attributes that make it suitable for portable data storage. Unlike RAM, flash memory is not erased when the device is powered down. Although there are other devices that provide this functionality, they have a number of downsides. ROM, or Read Only Memory, permanently stores the data with which it was imbued during its manufacture. For consumers who wish to store data at home, this isn't useful. Merely using a hard drive isn't useful, either, as it requires large amounts of power, contains moving parts and isn't nearly as portable.

Physical Mechanism

Transistors are responsible for the physical aspect of flash data storage. Most transistors can only store information when power is flowing through them. A component called the gate determines whether the transistor is in a state representing a "0" or a "1." Flash memory transistors have a second gate, called a floating gate. This traps electrons to maintain the transistor's value, even when no current is present.

Data Storage

Like most forms of storage, flash drives store memory using binary digits. A transistor can either store a "0" or a "1," based on the electrical charge present within it. If an electron is present within a gate, it is no longer conductive, producing a "0." A "1" is produced if there is no electron, thus making the transistor conductive. When a number of transistors are grouped together, this is called a block or sector. A series of these allows the computer to form complex data structures using the simple binary digits.

Wear Leveling

After a sufficient amount of usage, flash drive sectors go bad due to damage from the electric current. This can cause data loss if it is not pre-emptively addressed. To solve this problem, flash drive manufacturers employ a technique called wear levelling. Wear levelling causes the drive to store data in sectors which have been written to a relatively low number of times in an attempt to prevent sectors from wearing out quickly. A lookup table is stored on the drive so that the data is treated as if it were stored sequentially, even though it is not.

References

• Explain that Stuff!: How Flash Memory Works
• HITEQUEST: Introduction to Flash Memory
• Corsair: USB Flash Wear-Leveling and Life Span