Burn-in tests used and suggested

GPU standalone validation: GPU Validation Test

Symptoms observed through hardware burn-in testing

Component Type	Method	Fail condition	Other possible factors
Processor	mprime for Linux prime95 for Windows	System unexpectedly reboots, and reports multiple Memory DIMM slots CPU overheats Threshold: 80c-90c for most Broadwell Xeons; 75-80c for most Skylake/Scalable Idle: 45c-50c is high idle temps	Environment Debris/buildup on chassis Chassis fans
Memory	mprime for Linux (monitored through EDAC/IPMI) prime95 for Windows (monitored through 'system' event logs or IPMI)	System unexpectedly reboots, and offending Memory DIMM is logged in BMC (or Windows) event log
Graphics Card	Standalone Validation Test for Linux (log files with consistent/matching values) Windows 'binom' Tests (monitored through 'NVSMI' via CMD or event logs)	GPU will fall off BUS or report ERR! when monitoring 'nvidia-smi' (Linux); event may be reported in /var/log One or more GPU's significantly higher temp than others (92-95c) PCI-e device reported in Windows 'system' event log 'warnings'/'errors' (Windows)	Motherboard Processor
Power Supply	Running CPU, Memory, and GPU test at the same time	(Assuming correct outlet is being used) System unexpectedly powers off and does not reboot; logged in BMC event log
Motherboard	Critical motherboard issues will not POST or provide any display	No power LED displaying on board when power supply is known-working and plugged in

Component Type	Method	Fail condition	Other possible factors
Processor	mprime for Linux prime95 for Windows	System unexpectedly reboots; even under minimal POST configuration and cycled through all memory DIMM's CPU overheats Threshold: 80c-90c for most Broadwell Xeons; 75-80c for most Skylake/Scalable Idle: 45c-50c is high idle temps	Environment Debris/buildup on chassis CPU liquid cooler
Memory	mprime for Linux (monitored through EDAC/IPMI) prime95 for Windows (monitored through 'system' event logs or IPMI)	System unexpectedly reboots, and offending Memory DIMM might be logged in /var/log or Windows 'event logs'	Motherboard Processor
Graphics Card	Standalone Validation Test for Linux (log files with consistent/matching values) Windows 'binom' Tests (monitored through 'NVSMI' via CMD or event logs)	GPU will fall off BUS or report ERR! when monitoring 'nvidia-smi' (Linux); event may be reported in /var/log One or more GPU's significantly higher temp than others (92-95c) PCI-e device reported in Windows 'system' event log 'warnings'/'errors' (Windows)	Motherboard Processor
Power Supply	Running CPU, Memory, and GPU test at the same time	(Assuming correct outlet is being used) System unexpectedly powers off and does not reboot
Motherboard		No power LED displaying on board when power supply is known-working and plugged in