{
  "schema": "bounded-article-probe-evidence/v1",
  "date": "2026-11-04",
  "itemId": "SIG-20261104-006",
  "kind": "source-derived isolated module probe with synthetic inputs; no simulator/model/server execution",
  "originalExecutedAt": "2026-11-04T05:09:08.601403+00:00",
  "provenance": "Reused from completed same-date Analysis without new execution; source-derived synthetic inputs to the complete unmodified task metric module.",
  "sourcePins": [
    {
      "sha256": "1a27a261e543a201d17862d4d50393caf8751b0b072c2674f75cfb05fc7d48b3",
      "url": "https://raw.githubusercontent.com/ShengxuanQiu/MAServeBench/22483ab5df5eff562061953d564aac47a9423c2f/mas_workflow/app/semantic/metrics.py",
      "revision": "22483ab5df5eff562061953d564aac47a9423c2f",
      "originalAcquiredAt": "2026-11-04T03:16:31.197Z",
      "extractedFrom": "mas_workflow/app/semantic/metrics.py"
    }
  ],
  "records": [
    {
      "experiment": "EXP-006",
      "case": "baseline",
      "inputs": {
        "run_valid": true,
        "measurement_duration_sec": 20,
        "tasks": [
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 12
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 12
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 12
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 12
          }
        ]
      },
      "expected": {
        "attempted_tasks": 4,
        "slo_compliant_tasks": 0,
        "task_goodput": 0,
        "slo_attainment": 0,
        "task_latency_sec": {
          "p50": 12,
          "p90": 12,
          "p99": 12
        }
      },
      "actual": {
        "schema_version": "masbench.metrics/1.0.0",
        "run_valid": true,
        "task_latency_sec": {
          "p50": 12,
          "p90": 12,
          "p99": 12
        },
        "raw_wall_clock_task_latency_sec": {
          "p50": 12
        },
        "task_goodput": 0,
        "slo_attainment": 0,
        "attempted_tasks": 4,
        "successful_tasks": 4,
        "slo_compliant_tasks": 0,
        "slo_sec": 10,
        "resource_efficiency_tasks_per_sec_per_accelerator": 0,
        "sustainable_capacity": null,
        "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
      },
      "passed": true
    },
    {
      "experiment": "EXP-006",
      "case": "faster_but_late",
      "inputs": {
        "run_valid": true,
        "measurement_duration_sec": 20,
        "tasks": [
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 11,
            "task_latency_sec": 11
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 11,
            "task_latency_sec": 11
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 11,
            "task_latency_sec": 11
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 11,
            "task_latency_sec": 11
          }
        ]
      },
      "expected": {
        "slo_compliant_tasks": 0,
        "task_goodput": 0,
        "task_latency_sec": {
          "p50": 11,
          "p90": 11,
          "p99": 11
        }
      },
      "actual": {
        "schema_version": "masbench.metrics/1.0.0",
        "run_valid": true,
        "task_latency_sec": {
          "p50": 11,
          "p90": 11,
          "p99": 11
        },
        "raw_wall_clock_task_latency_sec": {
          "p50": 11
        },
        "task_goodput": 0,
        "slo_attainment": 0,
        "attempted_tasks": 4,
        "successful_tasks": 4,
        "slo_compliant_tasks": 0,
        "slo_sec": 10,
        "resource_efficiency_tasks_per_sec_per_accelerator": 0,
        "sustainable_capacity": null,
        "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
      },
      "passed": true
    },
    {
      "experiment": "EXP-006",
      "case": "deadline_boundary",
      "inputs": {
        "run_valid": true,
        "measurement_duration_sec": 20,
        "tasks": [
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 10,
            "task_latency_sec": 10
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 10,
            "task_latency_sec": 10
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 10,
            "task_latency_sec": 10
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 10,
            "task_latency_sec": 10
          }
        ]
      },
      "expected": {
        "slo_compliant_tasks": 4,
        "task_goodput": 0.2,
        "slo_attainment": 1
      },
      "actual": {
        "schema_version": "masbench.metrics/1.0.0",
        "run_valid": true,
        "task_latency_sec": {
          "p50": 10,
          "p90": 10,
          "p99": 10
        },
        "raw_wall_clock_task_latency_sec": {
          "p50": 10
        },
        "task_goodput": 0.2,
        "slo_attainment": 1,
        "attempted_tasks": 4,
        "successful_tasks": 4,
        "slo_compliant_tasks": 4,
        "slo_sec": 10,
        "resource_efficiency_tasks_per_sec_per_accelerator": 0.2,
        "sustainable_capacity": null,
        "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
      },
      "passed": true
    },
    {
      "experiment": "EXP-006",
      "case": "external_wait_control",
      "inputs": {
        "run_valid": true,
        "measurement_duration_sec": 20,
        "tasks": [
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 1
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 1
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 1
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 1
          }
        ]
      },
      "expected": {
        "slo_compliant_tasks": 0,
        "slo_attainment": 0
      },
      "actual": {
        "schema_version": "masbench.metrics/1.0.0",
        "run_valid": true,
        "task_latency_sec": {
          "p50": 12,
          "p90": 12,
          "p99": 12
        },
        "raw_wall_clock_task_latency_sec": {
          "p50": 12
        },
        "task_goodput": 0,
        "slo_attainment": 0,
        "attempted_tasks": 4,
        "successful_tasks": 4,
        "slo_compliant_tasks": 0,
        "slo_sec": 10,
        "resource_efficiency_tasks_per_sec_per_accelerator": 0,
        "sustainable_capacity": null,
        "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
      },
      "passed": true
    },
    {
      "experiment": "EXP-006",
      "case": "failed_and_missing_denominator",
      "inputs": {
        "run_valid": true,
        "measurement_duration_sec": 20,
        "tasks": [
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 5,
            "task_latency_sec": 5
          },
          {
            "successful": false,
            "raw_wall_clock_latency_sec": 6,
            "task_latency_sec": 6
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": null,
            "task_latency_sec": null
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 12
          }
        ]
      },
      "expected": {
        "attempted_tasks": 4,
        "successful_tasks": 3,
        "slo_compliant_tasks": 1,
        "slo_attainment": 0.25,
        "task_goodput": 0.05
      },
      "actual": {
        "schema_version": "masbench.metrics/1.0.0",
        "run_valid": true,
        "task_latency_sec": {
          "p50": 8.5,
          "p90": 11.3,
          "p99": 11.93
        },
        "raw_wall_clock_task_latency_sec": {
          "p50": 6
        },
        "task_goodput": 0.05,
        "slo_attainment": 0.25,
        "attempted_tasks": 4,
        "successful_tasks": 3,
        "slo_compliant_tasks": 1,
        "slo_sec": 10,
        "resource_efficiency_tasks_per_sec_per_accelerator": 0.05,
        "sustainable_capacity": null,
        "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
      },
      "passed": true
    },
    {
      "experiment": "EXP-006",
      "case": "survivor_before",
      "inputs": {
        "run_valid": true,
        "measurement_duration_sec": 20,
        "tasks": [
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 5,
            "task_latency_sec": 5
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 6,
            "task_latency_sec": 6
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 12
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 20,
            "task_latency_sec": 20
          }
        ]
      },
      "expected": {
        "slo_attainment": 0.5,
        "task_latency_sec": {
          "p50": 9,
          "p90": 17.6,
          "p99": 19.759999999999998
        }
      },
      "actual": {
        "schema_version": "masbench.metrics/1.0.0",
        "run_valid": true,
        "task_latency_sec": {
          "p50": 9,
          "p90": 17.6,
          "p99": 19.759999999999998
        },
        "raw_wall_clock_task_latency_sec": {
          "p50": 9
        },
        "task_goodput": 0.1,
        "slo_attainment": 0.5,
        "attempted_tasks": 4,
        "successful_tasks": 4,
        "slo_compliant_tasks": 2,
        "slo_sec": 10,
        "resource_efficiency_tasks_per_sec_per_accelerator": 0.1,
        "sustainable_capacity": null,
        "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
      },
      "passed": true
    },
    {
      "experiment": "EXP-006",
      "case": "survivor_after",
      "inputs": {
        "run_valid": true,
        "measurement_duration_sec": 20,
        "tasks": [
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 5,
            "task_latency_sec": 5
          },
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 6,
            "task_latency_sec": 6
          },
          {
            "successful": false,
            "raw_wall_clock_latency_sec": 12,
            "task_latency_sec": 12
          },
          {
            "successful": false,
            "raw_wall_clock_latency_sec": 20,
            "task_latency_sec": 20
          }
        ]
      },
      "expected": {
        "slo_attainment": 0.5,
        "task_latency_sec": {
          "p50": 5.5,
          "p90": 5.9,
          "p99": 5.99
        }
      },
      "actual": {
        "schema_version": "masbench.metrics/1.0.0",
        "run_valid": true,
        "task_latency_sec": {
          "p50": 5.5,
          "p90": 5.9,
          "p99": 5.989999999999999
        },
        "raw_wall_clock_task_latency_sec": {
          "p50": 9
        },
        "task_goodput": 0.1,
        "slo_attainment": 0.5,
        "attempted_tasks": 4,
        "successful_tasks": 2,
        "slo_compliant_tasks": 2,
        "slo_sec": 10,
        "resource_efficiency_tasks_per_sec_per_accelerator": 0.1,
        "sustainable_capacity": null,
        "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
      },
      "passed": true
    },
    {
      "experiment": "EXP-006",
      "case": "invalid_run_still_has_numbers",
      "inputs": {
        "run_valid": false,
        "measurement_duration_sec": 20,
        "tasks": [
          {
            "successful": true,
            "raw_wall_clock_latency_sec": 5,
            "task_latency_sec": 5
          }
        ]
      },
      "expected": {
        "run_valid": false,
        "task_goodput": 0.05,
        "slo_attainment": 1
      },
      "actual": {
        "schema_version": "masbench.metrics/1.0.0",
        "run_valid": false,
        "task_latency_sec": {
          "p50": 5,
          "p90": 5,
          "p99": 5
        },
        "raw_wall_clock_task_latency_sec": {
          "p50": 5
        },
        "task_goodput": 0.05,
        "slo_attainment": 1,
        "attempted_tasks": 1,
        "successful_tasks": 1,
        "slo_compliant_tasks": 1,
        "slo_sec": 10,
        "resource_efficiency_tasks_per_sec_per_accelerator": 0.05,
        "sustainable_capacity": null,
        "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
      },
      "passed": true
    },
    {
      "experiment": "EXP-006",
      "case": "capacity_validity_control",
      "inputs": [
        {
          "load": 1,
          "schema_version": "masbench.metrics/1.0.0",
          "run_valid": true,
          "task_latency_sec": {
            "p50": 5,
            "p90": 5,
            "p99": 5
          },
          "raw_wall_clock_task_latency_sec": {
            "p50": 5
          },
          "task_goodput": 0.05,
          "slo_attainment": 1,
          "attempted_tasks": 1,
          "successful_tasks": 1,
          "slo_compliant_tasks": 1,
          "slo_sec": 10,
          "resource_efficiency_tasks_per_sec_per_accelerator": 0.05,
          "sustainable_capacity": null,
          "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
        },
        {
          "load": 2,
          "schema_version": "masbench.metrics/1.0.0",
          "run_valid": false,
          "task_latency_sec": {
            "p50": 5,
            "p90": 5,
            "p99": 5
          },
          "raw_wall_clock_task_latency_sec": {
            "p50": 5
          },
          "task_goodput": 0.05,
          "slo_attainment": 1,
          "attempted_tasks": 1,
          "successful_tasks": 1,
          "slo_compliant_tasks": 1,
          "slo_sec": 10,
          "resource_efficiency_tasks_per_sec_per_accelerator": 0.05,
          "sustainable_capacity": null,
          "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
        },
        {
          "load": 3,
          "schema_version": "masbench.metrics/1.0.0",
          "run_valid": true,
          "task_latency_sec": {
            "p50": 12,
            "p90": 12,
            "p99": 12
          },
          "raw_wall_clock_task_latency_sec": {
            "p50": 12
          },
          "task_goodput": 0,
          "slo_attainment": 0,
          "attempted_tasks": 1,
          "successful_tasks": 1,
          "slo_compliant_tasks": 0,
          "slo_sec": 10,
          "resource_efficiency_tasks_per_sec_per_accelerator": 0,
          "sustainable_capacity": null,
          "sustainable_capacity_note": "Requires a configured multi-load sweep; use sustainable_capacity()."
        }
      ],
      "expected": {
        "maximum_tested_load": 1,
        "tested_loads": [
          1,
          2,
          3
        ]
      },
      "actual": {
        "attainment_target": 0.9,
        "maximum_tested_load": 1,
        "tested_loads": [
          1,
          2,
          3
        ]
      },
      "passed": true
    }
  ],
  "firstRunCorrection": {
    "cause": "Binary floating-point representation 5.99 versus 5.989999999999999",
    "change": "Only comparison tolerance adjusted in Analysis; original first run retained upstream."
  },
  "fullSystemReproduction": false,
  "independent": false
}