{
  "slug": "che-approximation",
  "term": "Che approximation",
  "aliases": [
    "Characteristic-time approximation",
    "TTL approximation for LRU"
  ],
  "definition": "The Che approximation estimates LRU hit probabilities by replacing random, object-dependent eviction times with a common deterministic characteristic time.",
  "note": "For independent Poisson requests with rates lambda_i and an equal-size, fully associative LRU cache of C objects, choose T so that sum_i (1 - exp(-lambda_i T)) = C. Then h_i = 1 - exp(-lambda_i T); request hit ratio is sum_i lambda_i h_i / sum_i lambda_i, not the unweighted mean. T is in seconds when rates are per second. This common-time form discards object-specific cutoffs. [fricker]\n\nWith two objects requested at rates 2 and 1 per second and one slot, Che predicts 53.93% hits; exact stationary LRU gives 55.56%. A one-slot hit requires consecutive requests to match, so its exact probability is (2/3)^2 + (1/3)^2 = 5/9. These are constructed model probabilities. [example]\n\nFor reset-on-request TTL with inter-request CDF F_i, occupancy is lambda_i times the integral of 1 - F_i(t) from 0 to T. Hit probability is F_i(T), away from expiry ties. Poisson arrivals make them coincide; general renewal arrivals need not. [renewal]\n\nAn object requested once per second with stationary random phase and a half-second TTL is resident half the time, but every request misses. This TTL example does not describe LRU. [example]\n\nThe linked code compares the common-time equation with exact stationary recency-order enumeration for small catalogues. A small numerical root residual does not bound the model's difference from LRU. [code]",
  "status": "research",
  "advantages": [],
  "limitations": [
    {
      "text": "Convergence requires growing caches and catalogues, independent stationary request processes and further regularity/popularity assumptions; it is not an exactness guarantee for a fixed small cache.",
      "citationIds": [
        "convergence"
      ]
    },
    {
      "text": "Variable-size byte-capacity equations additionally ignore object-boundary effects; correlated or changing requests and cache hierarchies require separate modelling.",
      "citationIds": [
        "fricker",
        "renewal"
      ]
    }
  ],
  "supersedes": [],
  "supersededBy": [],
  "firstSeen": null,
  "lastSeen": null,
  "updated": "2026-11-21",
  "related": [
    "independent-reference-model",
    "lru-replacement",
    "miss-ratio-curve",
    "average-eviction-time",
    "tsetlin-library"
  ],
  "citations": [
    {
      "id": "fricker",
      "title": "Fricker, Robert and Roberts. A versatile and accurate approximation for LRU cache performance. arXiv:1202.3974v1 (2012), sections 2-4.",
      "url": "https://arxiv.org/pdf/1202.3974v1",
      "accessed": "2026-11-19"
    },
    {
      "id": "renewal",
      "title": "Martina, Garetto and Leonardi. A unified approach to the performance analysis of caching systems. arXiv:1307.6702v6 (2016), sections II-IV.A.",
      "url": "https://arxiv.org/pdf/1307.6702v6",
      "accessed": "2026-11-19"
    },
    {
      "id": "convergence",
      "title": "Jiang, Nain and Towsley. On the Convergence of the TTL Approximation for an LRU Cache. arXiv:1707.06204v4 (2018), sections 2-4.",
      "url": "https://arxiv.org/pdf/1707.06204v4",
      "accessed": "2026-11-19"
    },
    {
      "id": "example",
      "title": "Hesela. Constructed common-time, exact-LRU and periodic-TTL controls, with explicit units and assumptions.",
      "url": "https://hesela.dev/research/che-approximation/evidence.json",
      "accessed": "2026-11-19"
    },
    {
      "id": "code",
      "title": "Hesela. Dependency-free arithmetic and exact small-catalogue stationary LRU enumeration.",
      "url": "https://hesela.dev/research/che-approximation/model.mjs",
      "accessed": "2026-11-19"
    },
    {
      "id": "tests",
      "title": "Hesela. Analytic, rate-scaling, monotonicity and invalid-input tests.",
      "url": "https://hesela.dev/research/che-approximation/model.test.mjs",
      "accessed": "2026-11-19"
    },
    {
      "id": "observatory",
      "title": "Hesela. Storage cache models and workload assumptions.",
      "url": "https://hesela.com/analyses/ssd-mapping-cache-hmb/",
      "accessed": "2026-11-19"
    }
  ],
  "provenance": [
    "Selected full-text methods and limitations read in three primary papers; no original workload or hardware benchmark reproduced.",
    "The name refers to Che, Tung and Wang (2002); the common-time equations here are verified from the explicitly versioned 2012 treatment, not a claimed reading of the original 2002 paper.",
    "Exact recency-order enumeration is Hesela's independent small-state arithmetic control, not the cited papers' experimental implementation."
  ]
}
