{
  "id": "dgxarley",
  "title": "DGXArley: cluster deployment and transport reports",
  "kind": "benchmark",
  "source_type": "community_operator_report",
  "summary": "An operator-maintained K3s/Ansible repository for a four-Spark cluster. It contains deployment notes and reported transport/inference comparisons. Select the exact model profile and experiment before reusing a number.",
  "runtime": "SGLang",
  "hardware_topics": [
    "spark"
  ],
  "platform_topics": [
    "linux"
  ],
  "model_label": "Multiple model profiles; see the selected report",
  "source": {
    "publisher": "vroomfondel",
    "url": "https://github.com/vroomfondel/dgxarley",
    "repository": "vroomfondel/dgxarley",
    "revision": "5f1f3762797eac0ea1ad43e403fa979e3e2c3934",
    "pinned_url": "https://github.com/vroomfondel/dgxarley/tree/5f1f3762797eac0ea1ad43e403fa979e3e2c3934",
    "last_commit_at": "2026-10-16T07:08:17Z",
    "checked_on": "2026-10-17",
    "metadata_status": "checked",
    "report_url": "https://github.com/vroomfondel/dgxarley/blob/5f1f3762797eac0ea1ad43e403fa979e3e2c3934/NCCL_BANDWIDTH_PROFILING.md",
    "platform": "github",
    "document_path": "NCCL_BANDWIDTH_PROFILING.md",
    "document_sha256": "293d119b6374d8073f040fe6dd277fe42f50f5f9baaadfdfa0d2120e65f12568"
  },
  "license": {
    "declared_spdx": "LGPL-3.0",
    "scope": "Repository declaration only; models, containers, datasets and per-file terms remain separate."
  },
  "evidence_state": "source_reported",
  "site_reproduced": false,
  "record_type": "curated_reference",
  "performance_ranking_allowed": false,
  "metrics": null,
  "metrics_status": "site_metrics_unmeasured; attributed_source_figures_in_editorial.reported_results",
  "conditions": {
    "model_revision": null,
    "weight_hash": null,
    "runtime_execution_revision": null,
    "input_tokens": null,
    "output_tokens": null,
    "concurrency": null,
    "cache_state": null
  },
  "notes": [
    "The repository contains multiple model profiles and parallelism choices.",
    "Transport, topology and concurrency belong to each reported experiment.",
    "Per-file licensing may differ from the repository’s main license declaration."
  ],
  "limitations": [
    "A default deployment profile is not necessarily the profile used in a benchmark paragraph.",
    "Prompt lengths, output policy, cache and model revision need verification in the underlying report.",
    "Do not compare these results directly with the three-node MiniMax configuration."
  ],
  "comparison_status": "not_comparable",
  "editorial_status": "source_derived_detailed_article",
  "public_benchmark_snapshot": false,
  "reported_setup": {
    "system": "NVIDIA DGX Spark / K3s cluster",
    "node_count": 4,
    "parallelism": null,
    "kv_dtype": null,
    "context_limit_tokens": null,
    "experiment_date": null
  },
  "runtime_topics": [
    "SGLang"
  ],
  "model_family": "Multiple",
  "model_id": null,
  "model_release_date": null,
  "model_artifact": null,
  "test_profiles": [],
  "deployment_options": [],
  "recipe_parameters": null,
  "measurement_scope": "transport_and_inference",
  "editorial": {
    "version": 1,
    "published_on": "2026-10-17",
    "modified_on": "2026-10-17",
    "byline": "Run AI Here editorial",
    "creation_method": "AI-assisted source reading and original editorial synthesis; no independent model execution.",
    "answer": "This document describes NCCL all-reduce profiling for a Spark serving cluster. It recommends inspecting small messages as well as the large-message bandwidth plateau. The author discusses roughly 185–190 versus 92–95 Gbit/s under different port-spanning conditions, but those are transport observations without an identified workload row here. They are not language-model decode speeds.",
    "sections": [
      {
        "heading": "What is the report measuring?",
        "paragraphs": [
          "This document describes NCCL all-reduce profiling for a Spark serving cluster. It recommends inspecting small messages as well as the large-message bandwidth plateau. The author discusses roughly 185–190 versus 92–95 Gbit/s under different port-spanning conditions, but those are transport observations without an identified workload row here. They are not language-model decode speeds."
        ],
        "citations": [
          "https://github.com/vroomfondel/dgxarley/blob/5f1f3762797eac0ea1ad43e403fa979e3e2c3934/NCCL_BANDWIDTH_PROFILING.md#L12"
        ]
      },
      {
        "heading": "Why is a healthy headline link rate insufficient?",
        "paragraphs": [
          "A standalone benchmark can inherit different interface, channel or RoCE settings from the serving container. Large messages can also hide the latency-bound region that tensor-parallel decode uses. Record the actual serving path and inspect its small-message behavior before attributing an inference slowdown to model kernels."
        ],
        "citations": [
          "https://github.com/vroomfondel/dgxarley/blob/5f1f3762797eac0ea1ad43e403fa979e3e2c3934/NCCL_BANDWIDTH_PROFILING.md#L25"
        ]
      },
      {
        "heading": "How is the sweep defined?",
        "paragraphs": [
          "The tool’s default range starts at 1 MB and ends at 1 GB, with a factor-two size step, five warmup and twenty timed iterations per size. The document asks operators to lower the floor to 8 KB for decode-oriented profiling. The ratio below 0.4 is an author-defined diagnostic flag, not an industry pass/fail threshold or a prediction of tokens per second."
        ],
        "citations": [
          "https://github.com/vroomfondel/dgxarley/blob/5f1f3762797eac0ea1ad43e403fa979e3e2c3934/NCCL_BANDWIDTH_PROFILING.md#L49"
        ]
      },
      {
        "heading": "What belongs in a usable transport receipt?",
        "paragraphs": [
          "Keep node count, per-node network device, MTU path, transport, process/container network settings, message size and both algorithm and bus bandwidth. Match those settings to the inference process before linking network results to generation performance. Maintenance and live-workload conflicts require the operator’s own planning; this site does not inspect or alter any cluster."
        ],
        "citations": [
          "https://github.com/vroomfondel/dgxarley/blob/5f1f3762797eac0ea1ad43e403fa979e3e2c3934/NCCL_BANDWIDTH_PROFILING.md#L130"
        ]
      }
    ],
    "facts": [
      {
        "name": "Decode-oriented message region",
        "value": "8 KB–1 MB",
        "classification": "publisher_declaration",
        "evidence_url": "https://github.com/vroomfondel/dgxarley/blob/5f1f3762797eac0ea1ad43e403fa979e3e2c3934/NCCL_BANDWIDTH_PROFILING.md#L33",
        "source_document_sha256": "293d119b6374d8073f040fe6dd277fe42f50f5f9baaadfdfa0d2120e65f12568"
      },
      {
        "name": "Default warmup / timed iterations",
        "value": "5 / 20",
        "classification": "publisher_declaration",
        "evidence_url": "https://github.com/vroomfondel/dgxarley/blob/5f1f3762797eac0ea1ad43e403fa979e3e2c3934/NCCL_BANDWIDTH_PROFILING.md#L52",
        "source_document_sha256": "293d119b6374d8073f040fe6dd277fe42f50f5f9baaadfdfa0d2120e65f12568"
      },
      {
        "name": "Author’s small-region warning threshold",
        "value": "ratio below 0.4× overall peak",
        "classification": "publisher_declaration",
        "evidence_url": "https://github.com/vroomfondel/dgxarley/blob/5f1f3762797eac0ea1ad43e403fa979e3e2c3934/NCCL_BANDWIDTH_PROFILING.md#L59",
        "source_document_sha256": "293d119b6374d8073f040fe6dd277fe42f50f5f9baaadfdfa0d2120e65f12568"
      }
    ],
    "reported_results": [],
    "source_document_sha256": "293d119b6374d8073f040fe6dd277fe42f50f5f9baaadfdfa0d2120e65f12568"
  }
}