{
  "abstract": "Repeatedly sampling short sequence windows from a human pangenome requires an access representation as well as compression. panvram stores each assembly as a reference-relative block archive and pools these archives on a GPU. For 558 haplotype assemblies, the default q4k representation occupies 16.642 GB of persistent pools including the decoded reference; a packed-reference, compact-directory representation occupies 11.634 GB. These quantities are not peak application memory. In the C1 experiment, default and fully compact forms delivered 743,271 and 736,762 windows/s; a separate public Colab run delivered 741,931 and 680,760 windows/s. All are batch API measurements and must not be pooled as repeated trials of an identical environment. Full-decoding checks compared reconstructed FASTA checksums with the manifest. On the CPU, the M6 shuffled-window workload at the same cohort cardinality reported 26 requests/s for AGC and 145,222 requests/s for refrel3 q4k. This comparison concerns one-thread requests of 4 KiB on the recorded warm-cache workload, not all genome-access patterns. AGC provides the smaller incremental archive: 5.40 MB/assembly versus 13.04 MB/assembly for refrel3 q16k, with different exclusions; the size pair is not the same refrel3 configuration as the q4k timing pair. We derive an expected-work expression for fixed independent granules and distinguish it from an empirical median-latency predictor. Cohort-scale counter and DRAM measurements motivate a locality hypothesis, but do not establish a causal per-miss latency or independently reproduce a model-failure verdict. The window-cost verdict at N=558 is published with its harness and raw timings.",
  "abstract_shown": true,
  "affiliations": [],
  "article_type": null,
  "arxiv": "2610.25149",
  "author_total": 1,
  "authors": [
    {
      "name": "Yakiv Shavidze"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/a3qahepqck760.bib",
  "cited_by": 0,
  "doi": "10.48550/arxiv.2610.25149",
  "fulltext_url": null,
  "id": "a3qahepqck760",
  "issns": [],
  "landing_url": "https://arxiv.org/abs/2610.25149",
  "language": null,
  "licence": "CC-BY-4.0",
  "licence_label": "CC BY 4.0",
  "licence_source": "article",
  "licence_url": "http://creativecommons.org/licenses/by/4.0/",
  "metadata_licence": "CC0-1.0",
  "native_subjects": [
    [
      "arxiv",
      "cs.DC"
    ],
    [
      "arxiv",
      "cs.PF"
    ],
    [
      "arxiv",
      "q-bio.GN"
    ],
    [
      "openalex",
      "1702|Algorithms and Data Compression"
    ]
  ],
  "openalex": "W7223052544",
  "pdf_url": "https://arxiv.org/pdf/2610.25149",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-18",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/a3qahepqck760.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:31:01Z",
      "native_id": "2610.25149",
      "source": "arxiv",
      "url": "https://arxiv.org/abs/2610.25149"
    }
  ],
  "subjects": [
    {
      "field": "biology",
      "origin": "arxiv:q-bio.GN",
      "subfield": "genetics-genomics"
    }
  ],
  "title": "panvram: GPU-resident random access to a human pangenome cohort",
  "title_text": "panvram: GPU-resident random access to a human pangenome cohort",
  "type": "preprint",
  "updated_at": "2026-10-30T12:33:50Z",
  "url": "https://research.draconitic.com/paper/a3qahepqck760",
  "venue": "arXiv",
  "versions": []
}