Documentation
¶
Index ¶
Constants ¶
This section is empty.
Variables ¶
This section is empty.
Functions ¶
This section is empty.
Types ¶
type GPUHealthChecker ¶
type GPUHealthChecker struct {
// contains filtered or unexported fields
}
GPUHealthChecker checks the health of nvidia GPUs. Note that with the current device naming pattern in device manager, GPUHealthChecker will not work with MIG devices.
func NewGPUHealthChecker ¶
func NewGPUHealthChecker(devices map[string]pluginapi.Device, health chan pluginapi.Device, codes []int) *GPUHealthChecker
NewGPUHealthChecker returns a GPUHealthChecker object for a given device name
func (*GPUHealthChecker) Start ¶
func (hc *GPUHealthChecker) Start() error
Start registers NVML events and starts listening to them
func (*GPUHealthChecker) Stop ¶
func (hc *GPUHealthChecker) Stop()
Stop deletes the NVML events and stops the listening go routine
Source Files
Click to show internal directories.
Click to hide internal directories.