{
  "slug": "stow",
  "term": "STOW",
  "aliases": [
    "Spatially and Temporally Optimized Writes"
  ],
  "definition": "STOW is a write-cache policy that separates sequential and random write groups and adapts their destage order, concurrency and cache allocation.",
  "note": "Two address-sorted queues retain WOW-style reference bits. Occupancy thresholds control concurrent destages; hysteresis batches work from one queue to limit switching. [stow-paper]\n\nThe batch ends after a page-count threshold or sufficient growth in either queue. The count is pages, not time or write groups. [stow-paper]\n\nThe linked RAID analysis provides a switching-cost example with explicit assumptions. [hesela-raid]",
  "status": "research",
  "advantages": [],
  "limitations": [
    {
      "text": "The rotating-disk evaluation used volatile DDR2 cache and an SPC-1 Like prototype workload, not a power-loss survival test.",
      "citationIds": [
        "stow-paper"
      ]
    }
  ],
  "related": [
    "wise-ordering-for-writes",
    "write-back-cache",
    "write-combining",
    "read-modify-write"
  ],
  "supersedes": [],
  "supersededBy": [],
  "firstSeen": null,
  "lastSeen": null,
  "citations": [
    {
      "id": "stow-paper",
      "title": "Gill, Ko, Debnath and Belluomini. STOW. USENIX ATC 2009, sections 3-6 and Algorithms 1-3.",
      "url": "https://www.usenix.org/legacy/event/usenix09/tech/full_papers/gill/gill.pdf",
      "accessed": "2026-11-21"
    },
    {
      "id": "hesela-raid",
      "title": "Hesela: RAID write penalty is not a speed ratio, destage batching example.",
      "url": "https://hesela.com/analyses/raid-write-penalty-io-not-speed/#stow-destaging",
      "accessed": "2026-11-21"
    }
  ],
  "provenance": [
    "Primary-source review of the algorithm and evaluation. The linked arithmetic is a synthetic scheduling illustration, not a STOW implementation or replicated benchmark."
  ],
  "updated": "2026-11-21"
}
