nvlink

Documentation

Overview

Package nvlink provides the NVIDIA nvlink metrics collection and reporting.

Index

• Constants
• func InitAveragers(db *sql.DB, tableName string)
• func ReadCRCErrors(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)
• func ReadFeatureEnabled(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)
• func ReadRecoveryErrors(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)
• func ReadReplayErrors(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)
• func ReadRxBytes(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)
• func ReadTxBytes(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)
• func Register(reg *prometheus.Registry, db *sql.DB, tableName string) error
• func SetCRCErrors(ctx context.Context, gpuID string, errors uint64, currentTime time.Time) error
• func SetFeatureEnabled(ctx context.Context, gpuID string, enabled bool, currentTime time.Time) error
• func SetLastUpdateUnixSeconds(unixSeconds float64)
• func SetRecoveryErrors(ctx context.Context, gpuID string, errors uint64, currentTime time.Time) error
• func SetReplayErrors(ctx context.Context, gpuID string, errors uint64, currentTime time.Time) error
• func SetRxBytes(ctx context.Context, gpuID string, bytes float64, currentTime time.Time) error
• func SetTxBytes(ctx context.Context, gpuID string, bytes float64, currentTime time.Time) error

Constants

const SubSystem = "accelerator_nvidia_nvlink"

Functions

func InitAveragers

func InitAveragers(db *sql.DB, tableName string)

func ReadCRCErrors

func ReadCRCErrors(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)

func ReadFeatureEnabled

func ReadFeatureEnabled(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)

func ReadRecoveryErrors

func ReadRecoveryErrors(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)

func ReadReplayErrors

func ReadReplayErrors(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)

func ReadRxBytes

func ReadRxBytes(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)

func ReadTxBytes

func ReadTxBytes(ctx context.Context, since time.Time) (components_metrics_state.Metrics, error)

func Register

func Register(reg *prometheus.Registry, db *sql.DB, tableName string) error

func SetCRCErrors

func SetCRCErrors(ctx context.Context, gpuID string, errors uint64, currentTime time.Time) error

func SetFeatureEnabled

func SetFeatureEnabled(ctx context.Context, gpuID string, enabled bool, currentTime time.Time) error

func SetLastUpdateUnixSeconds

func SetLastUpdateUnixSeconds(unixSeconds float64)

func SetRecoveryErrors

func SetRecoveryErrors(ctx context.Context, gpuID string, errors uint64, currentTime time.Time) error

func SetReplayErrors

func SetReplayErrors(ctx context.Context, gpuID string, errors uint64, currentTime time.Time) error

func SetRxBytes

func SetRxBytes(ctx context.Context, gpuID string, bytes float64, currentTime time.Time) error

func SetTxBytes

func SetTxBytes(ctx context.Context, gpuID string, bytes float64, currentTime time.Time) error