RAID (redundant array of independent disks)

Spreads data, with mirror copies or parity, across many cheap disks so the set is faster than one big drive and survives a disk failure; the basis of server and cloud storage.

Memory & storageHardware
  1. 1988

    First version

    David Patterson, Garth Gibson and Randy Katz (University of California, Berkeley)

    Image: Erik Pitti from San Diego, CA, USA · CC BY 2.0 · Wikimedia Commons
  2. 1990
    EMC Symmetrix cached disk array

    EMC Symmetrix cached disk array

    Arrays of small disks with a large cache stand in for IBM's big mainframe drives and win much of that market.

    Image: Bubba73 · CC BY-SA 3.0 · Wikimedia Commons
  3. 2005

    RAID-Z in ZFS

    Software RAID with end-to-end checksums and copy-on-write closes the 'write hole' and catches silent data corruption.

    Image: ™/®OpenZFS · Public domain · Wikimedia Commons
  4. 2012

    Erasure coding at cloud scale

    Windows Azure Storage spreads coded fragments across servers with Local Reconstruction Codes, cutting storage overhead and repair traffic.

    Image: Microsoft Corporation · Public domain · Wikimedia Commons

Contested: The ideas predate the name: Tandem and DEC mirrored disks in the 1970s and early 1980s, and IBM's Norman Ken Ouchi filed a parity-disk patent in 1977 (granted 1978). The 1988 Berkeley paper named and classified the RAID levels.

See it on the timeline