worker

package

v1.4.0 Latest Latest Go to latest Published: Feb 11, 2026 License: Apache-2.0 Imports: 13 Imported by: 0

Details

Valid go.mod file
Redistributable license
Tagged version
Stable version
Learn more about best practices

Repository

github.com/NexusGPU/gpu-go

Links

Open Source Insights

Documentation ¶

Rendered for

Index ¶

Constants
type Manager
- func NewManager(stateDir, workerBinary string) *Manager
type TensorFusionWorkerInfo
type WorkerConfig
type WorkerMode
type WorkerState
type WorkerStatus

Constants ¶

View Source

const DefaultWorkerPort = 42352

DefaultWorkerPort returns the default worker port

Variables ¶

This section is empty.

Functions ¶

This section is empty.

Types ¶

type Manager ¶

type Manager struct {
	// contains filtered or unexported fields
}

Manager manages worker processes

func NewManager ¶

func NewManager(stateDir, workerBinary string) *Manager

NewManager creates a new worker manager

func (*Manager) GetStatus ¶

func (m *Manager) GetStatus(workerID string) (*WorkerState, error)

GetStatus returns the status of a worker

func (*Manager) List ¶

func (m *Manager) List() []*WorkerState

List returns all worker states

func (*Manager) LoadStateFile ¶

func (m *Manager) LoadStateFile() ([]WorkerConfig, error)

LoadStateFile loads worker state from tensor-fusion format file

func (*Manager) Reconcile ¶

func (m *Manager) Reconcile(desiredWorkers []WorkerConfig) error

Reconcile reconciles the desired worker state with actual state

func (*Manager) SaveStateFile ¶

func (m *Manager) SaveStateFile() error

SaveStateFile saves worker state to a file in tensor-fusion format

func (*Manager) Shutdown ¶

func (m *Manager) Shutdown()

Shutdown stops all workers and cleans up

func (*Manager) Start ¶

func (m *Manager) Start(config WorkerConfig) error

Start starts a worker process

func (*Manager) Stop ¶

func (m *Manager) Stop(workerID string) error

Stop stops a worker process

type TensorFusionWorkerInfo ¶

type TensorFusionWorkerInfo struct {
	WorkerUID        string   `json:"WorkerUID"`
	Namespace        string   `json:"Namespace,omitempty"`
	WorkerName       string   `json:"WorkerName,omitempty"`
	AllocatedDevices []string `json:"AllocatedDevices"`
	Status           string   `json:"Status"` // "Pending", "Running", "Terminated"
}

TensorFusionWorkerInfo represents the worker info format expected by tensor-fusion hypervisor

type WorkerConfig ¶

type WorkerConfig struct {
	WorkerID         string     `json:"worker_id"`
	GPUIDs           []string   `json:"gpu_ids"`
	ListenPort       int        `json:"listen_port"`
	Mode             WorkerMode `json:"mode"`
	ShmemFile        string     `json:"shmem_file,omitempty"`
	ShmemSizeMB      int        `json:"shmem_size_mb,omitempty"`
	Enabled          bool       `json:"enabled"`
	WorkerBinaryPath string     `json:"worker_binary_path,omitempty"`
}

WorkerConfig represents configuration for a worker process

type WorkerMode ¶

type WorkerMode string

WorkerMode represents the worker network mode

const (
	WorkerModeTCP   WorkerMode = "tcp"
	WorkerModeShmem WorkerMode = "shmem"
)

type WorkerState ¶

type WorkerState struct {
	Config    WorkerConfig `json:"config"`
	Status    WorkerStatus `json:"status"`
	PID       int          `json:"pid,omitempty"`
	StartedAt *time.Time   `json:"started_at,omitempty"`
	Error     string       `json:"error,omitempty"`
}

WorkerState represents the runtime state of a worker

type WorkerStatus ¶

type WorkerStatus string

WorkerStatus represents the current status of a worker

const (
	WorkerStatusPending    WorkerStatus = "Pending"
	WorkerStatusRunning    WorkerStatus = "Running"
	WorkerStatusStopping   WorkerStatus = "Stopping"
	WorkerStatusStopped    WorkerStatus = "Stopped"
	WorkerStatusTerminated WorkerStatus = "Terminated"
	WorkerStatusError      WorkerStatus = "Error"
)

Source Files ¶

View all Source files

?	: This menu
/	: Search site
f or F	: Jump to
y or Y	: Canonical URL