Dynamic RAM

From Computer History Wiki
Revision as of 01:30, 17 February 2024 by Jnc (talk | contribs) (Explain why DRAM is such a low-power approach)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Dynamic RAM ('dynamic random-access memory'), usually abbreviated as DRAM, is currently the ubiquitous memory technology used for main memory in computers. Contemporary DRAM is implemented in integrated circuits, but very early DRAM implementations used discrete components. (It is actually a very old idea; the very first digital electronic computing device, the Atanasoff-Berry Computer, used a form of DRAM for its primary memory.)

It consists of an array of small capacitors; the presence, or absence, of a charge (stored electrons) in a particular capacitor represents the storage of a '1' or '0' bit. Over time, the charge dissipates as the electrons slowly leak away, so it must be refreshed on a regular basis; typically in the low numbers of milliseconds in contemporary DRAM.

Alternative semiconductor memory approaches, such as SRAM, use current flows to store information; the use of charge, stored in capacitors, allows DRAM to use a lot less power (which inevitably creates heat, which must be dealt with). DRAM only uses power when information is being read or written (which necessarily includes refresh operations).

DRAM is also volatile; it loses its contents if power is cut off.

External links

  • How we made DRAM - wonderful essay from Robert H. Dennard, about how he had the idea which led to IBM's 1968 patent on single-transistor dynamic random access memory