Flash memory is now mainstream, and is not just limited to removable solid-state storage. The original Flash memory used NOR gates, but most Flash memory uses NAND gates now as they are cheaper. The NAND gate IO interface handles blocks of data, which makes it an ideal substitute for data that was traditionally held on disk. Flash is only one type of solid state memory, the various types of solid-state storage summarised in the table below. The ideal memory subsystem is high capacity, retains data indefinitely, is fast for both write, and re-write, can be read fast, and is cost-effective for the application.
Memory Type
Cost
Density
Speed
Volatility
FLASH
Low-cost
High density
Fast read
10 year data life, non-volatile, re-writeable
ROM Read-Only Memory
Low cost
High density
Fast read
Mature and stable, data written during manufacture
SRAM Static Random-Access Memory
Expensive
Low density
Fastest speed
Data volatile, needs battery backup
EPROM Electrically Programmable Read-Only Memory
Medium cost
High-density
Fast
Will survive power loss, must be exposed to ultraviolet light to rewrite data.
EEPROM or E2 Electrically Erasable Programmable Read-Only Memory
High cost
lowest density
Fast
Data can be re-written electrically, data retention less reliable
DRAM Dynamic Random Access Memory
Low-cost
High- density
High-speed
High-power requirement
How does Flash Memory work?
A Flash memory cell is effectively a transistor. The original flash technology used NOR devices, although most devices now use NAND technology that are more reliable and have a faster re-write speed.
On NOR devices, the data charge is stored in a floating gate that is sandwiched between two insulating silicon dioxide layers that
in turn are sandwiched between the control gate and the transistor base. The memory is segmented into erase blocks, so whole blocks
of data must be erased at the same time, rather than individual bytes.
The diagram shows an Intel FLASH memory cell. Intel calls can exist in two states, '0' or'1', or in four states, '00' '01' '10' or '11'. This is achieved by precisely measuring the charge level, so 4 levels of change are recognised
Applications of Flash Storage
Flash Storage used to be limited to portable devices that store relatively small amounts of data. Examples include mobile phone cards, digital camera memories and USB sticks. However, Flash storage is now being used in the enterprise world, as a top tier for EMC DMX-4 devices, and as the primary storage disks in some Netbook computers. It is still quite a bit more expensive than hard disk drives and is not expected to replace them. However that may change, as Texas Memory Systems have released a rack mounted Flash enclosure that can hold up to 5 TB of capacity.
The best of both worlds
Flash, or NAND memory is relatively slow compared to the DRAM memory used in server and PC memory. DRAM is fast, but the data is volatile, and is lost when the power drops. Imagine the result if you combined the two technologies, very fast access combined with non-volatile data.
Well that is exactly what DDRdrive have done with their DDR Drive X1, released in early May, 2009. The initial device contains 4 GB of DRAM backed up by 4 GB of NAND Flash memory. Your applications interface with the very fast DRAM storage, while updates are written out to the NAND flash memory for safekeeping. The asking price is $1,495.00 with a 5 Year Limited Warranty.
Christopher George, the CEO and founder of DDRdrive, states that the DDRdrive X1 is "The drive for speed" (a trademarked description) and should be targeted at very IO intensive applications like databases, while applications that do not need blinding performance can be kept on spinning disk.