{
  "slug": "learnedftl",
  "term": "LearnedFTL",
  "aliases": [],
  "definition": "LearnedFTL is a research page-level flash translation layer that supplements a mapping cache with bitmap-qualified address prediction.",
  "note": "After a cached-mapping miss, a per-logical-page bitmap selects either a model-derived address or the translation-page fallback. A set bit marks an exact prediction for that mapping, not permission to read approximate data. The paper clears the bit on an overwrite. [paper]\n\nVirtual physical page numbers make parallel placement easier to model. Group-based allocation brings related mappings together for training during garbage collection. [paper]\n\nIn the pinned author artifact, model_predict compares its prediction with the emulator's mapping table, while the bitmap-clearing block in ssd_write is commented out. In ssd_read, model_hit_num increases before model eligibility and prediction success are checked; a failed prediction can still reach the translation-page fallback. That counter therefore does not, by itself, count successful avoided translation reads. This is a static observation of the pinned source, not a demonstrated data-loss event or a reproduced performance result. [code]\n\nLeaFTL is a different design. Its bounded-error segments use out-of-band metadata to correct mispredictions. Similar names do not make the algorithms or evaluation artifacts interchangeable. [lea]\n\nThe HPCA evaluation uses FEMU and reimplemented comparison algorithms. Retention-rewrite training is proposed but not evaluated there: the emulator lacks that rewrite mechanism, so the evaluated model training uses garbage collection. [paper]",
  "status": "research",
  "advantages": [
    {
      "text": "A qualified prediction can avoid a mapping-page read even when the mapping cache misses.",
      "citationIds": [
        "paper"
      ]
    }
  ],
  "limitations": [
    {
      "text": "A historical accuracy bit must remain consistent with the current mapping; prediction coverage is not a crash-recovery guarantee.",
      "citationIds": [
        "paper",
        "code"
      ]
    },
    {
      "text": "The artifact's setup requires a virtual machine and substantial host memory.",
      "citationIds": [
        "artifact"
      ]
    }
  ],
  "related": [
    "mapping-cache",
    "leaftl",
    "dftl",
    "global-translation-directory",
    "translation-page",
    "ftl",
    "garbage-collection"
  ],
  "supersedes": [],
  "supersededBy": [],
  "firstSeen": null,
  "lastSeen": null,
  "updated": "2026-11-22",
  "citations": [
    {
      "id": "paper",
      "title": "Wang et al. LearnedFTL: A Learning-Based Page-Level FTL for Reducing Double Reads in Flash-Based SSDs. HPCA 2024, sections III-IV.",
      "url": "https://ranger.uta.edu/~jiang/publication/Conferences/2024/HPCA24%28LearnedFTL%29.pdf",
      "accessed": "2026-11-22"
    },
    {
      "id": "code",
      "title": "LearnedFTL author artifact, commit 4702166cdc8daf32008a65a79f72ad991c2a991d: model_predict, ssd_read and ssd_write.",
      "url": "https://github.com/astlxmu/LearnedFTL/blob/4702166cdc8daf32008a65a79f72ad991c2a991d/hw/femu/bbssd/ld-tpftl.c",
      "accessed": "2026-11-22"
    },
    {
      "id": "artifact",
      "title": "LearnedFTL author artifact: build, virtual-machine setup and FTL selection at the inspected commit.",
      "url": "https://github.com/astlxmu/LearnedFTL/blob/4702166cdc8daf32008a65a79f72ad991c2a991d/README.md",
      "accessed": "2026-11-22"
    },
    {
      "id": "lea",
      "title": "Sun et al. LeaFTL: A Learning-Based Flash Translation Layer for Solid-State Drives. ASPLOS 2023, section 3.",
      "url": "https://jianh.web.engr.illinois.edu/papers/leaftl-asplos2023.pdf",
      "accessed": "2026-11-22"
    },
    {
      "id": "analysis",
      "title": "Hesela: SSD mapping caches, learned lookup validation and evaluation boundaries.",
      "url": "https://hesela.com/analyses/ssd-mapping-cache-hmb/#research",
      "accessed": "2026-11-22"
    }
  ],
  "provenance": [
    "Primary-paper synthesis and a static review of the pinned author implementation. No emulator run, hardware test, benchmark replication or independent review is claimed.",
    "The counter interpretation follows source control flow; it does not establish the numerical impact on published graphs. Paper design and inspected artifact are distinguished."
  ]
}
