{
  "slug": "harmonic-mean",
  "term": "Harmonic mean",
  "aliases": [],
  "definition": "The harmonic mean of positive values is their count divided by the sum of their reciprocals.",
  "note": "For n positive values x_i, H = n / sum(1/x_i); the result has the same units as the observations. A weighted harmonic mean is sum(w_i) / sum(w_i/x_i) for positive weights. State the population and weights rather than choosing a mean from the unit alone. [nist-harmonic] [python-harmonic]\n\nFor sequential, nonoverlapping transfers with matching byte accounting, aggregate throughput is total bytes / total elapsed time. Equal byte volumes make this the unweighted harmonic mean of phase rates; unequal volumes require byte weights. Equal durations instead give the arithmetic mean. Concurrent jobs require their common wall-clock interval, not summed overlapping durations. [python-harmonic] [hesela-means]\n\nIn a constructed example, 1000 MB at 100 MB/s takes 10 s and 1000 MB at 400 MB/s takes 2.5 s: 2000 MB / 12.5 s = 160 MB/s. Ten seconds at each rate instead gives 5000 MB / 20 s = 250 MB/s. MB means 1000000 bytes; these are accounting assumptions, not measured SSD performance. Include stalls in the elapsed interval when reporting end-to-end throughput. [hesela-means]",
  "related": [
    "geometric-mean",
    "throughput-storage"
  ],
  "citations": [
    {
      "id": "nist-harmonic",
      "title": "NIST Dataplot: Harmonic Mean, reciprocal-mean definition",
      "url": "https://itl.nist.gov/div898/software/dataplot/refman2/auxillar/harmmean.htm",
      "accessed": "2026-11-19"
    },
    {
      "id": "python-harmonic",
      "title": "Python statistics: harmonic_mean, equal-distance and weighted rate examples",
      "url": "https://docs.python.org/3/library/statistics.html#statistics.harmonic_mean",
      "accessed": "2026-11-19"
    },
    {
      "id": "hesela-means",
      "title": "Hesela: exact synthetic rate and benchmark-ratio examples, executable module",
      "url": "https://hesela.com/research/means/model.mjs",
      "accessed": "2026-11-19"
    }
  ],
  "status": "research",
  "advantages": [],
  "limitations": [],
  "supersedes": [],
  "supersededBy": [],
  "firstSeen": null,
  "lastSeen": null,
  "updated": "2026-11-19"
}
