stream-render-farm/fault-tree-analysis.json

61 lines
2.0 KiB
JSON
Raw Permalink Normal View History

2026-07-18 21:32:55 +02:00
{
"title": "Fault Tree Analysis for Render Farm Cluster Failure",
"slug": "fault-tree-analysis",
"wikidata_entities": {
"fault_tree_analysis": "Q428453",
"render_farm": "Q382597"
},
"top_event": {
"id": "CLUSTER_FAILURE",
"condition": "≥50% of render nodes offline simultaneously",
"gate_type": "OR"
},
"gates": [
{
"id": "POWER_FAILURE",
"gate_type": "AND",
"children": ["PSU_A_FAIL", "PSU_B_FAIL"]
},
{
"id": "COOLING_FAILURE",
"gate_type": "OR",
"children": ["CRAC_UNIT_DOWN", "HOTSPOT_TEMP_EXCEEDED"]
},
{
"id": "NETWORK_PARTITION",
"gate_type": "AND",
"children": ["SWITCH_A_DOWN", "SWITCH_B_DOWN"]
},
{
"id": "SOFTWARE_CASCADE",
"gate_type": "OR",
"children": ["DRIVER_CRASH", "QUEUE_MANAGER_HANG"]
}
],
"basic_events": [
{"id": "PSU_A_FAIL", "probability": 0.001, "mitigation": "hot-swap PSU"},
{"id": "PSU_B_FAIL", "probability": 0.001, "mitigation": "hot-swap PSU"},
{"id": "CRAC_UNIT_DOWN", "probability": 0.002, "mitigation": "redundant CRAC"},
{"id": "HOTSPOT_TEMP_EXCEEDED", "probability": 0.01, "threshold": "85°C", "mitigation": "active airflow redistribution"},
{"id": "SWITCH_A_DOWN", "probability": 0.0005, "mitigation": "stackwise redundancy"},
{"id": "SWITCH_B_DOWN", "probability": 0.0005, "mitigation": "stackwise redundancy"},
{"id": "DRIVER_CRASH", "probability": 0.05, "mitigation": "driver pinning + rollback"},
{"id": "QUEUE_MANAGER_HANG", "probability": 0.03, "mitigation": "watchdog timer + auto-reset"}
],
"minimal_cut_sets": [
["PSU_A_FAIL", "PSU_B_FAIL"],
["CRAC_UNIT_DOWN"],
["HOTSPOT_TEMP_EXCEEDED"],
["SWITCH_A_DOWN", "SWITCH_B_DOWN"],
["DRIVER_CRASH"],
["QUEUE_MANAGER_HANG"]
],
"probability_formula": {
"OR_gate": "P(TOP) = 1 ∏(1P_i)",
"AND_gate": "P(TOP) = ∏P_i"
},
"version": "1.0",
"published": "2026-07-18T16:32Z",
"author": "stream.4ort.net"
}