百度智能云

All Product Document

          Cloud Monitor

          Baidu Cloud Compute (BCC)

          Baidu Cloud Compute (BCC) only includes one type of monitor object: instance monitor (Instance). The list of monitoring metrics for instance monitor is as follows:

          Instance Monitor (instance)

          Metric name (English) Metric name (Chinese) Unit Dimension Remarks
          CpuIdlePercent CPU idle ratio % InstanceId
          CpuLoadAvg1 Server load within the last 1 minute Item InstanceId Exclusive to Linux server
          CpuLoadAvg15 Server load within the last 15 minutes Item InstanceId Exclusive to Linux server
          CpuLoadAvg5 Server load within the last 5 minutes Item InstanceId Exclusive to Linux server
          CpuSystemPercent System CPU time ratio % InstanceId
          CpuUserPercent User CPU time ratio % InstanceId
          CpuWaitPercent CPU IO-wait time ratio % InstanceId Exclusive to Linux server
          Cpu0IdlePercent Single-core CPU idle rate % InstanceId Exclusive to Windows server
          Cpu0ProcessorPercent Single-core CPU utilization % InstanceId Exclusive to Windows server
          CpuContextSwitchSecond Context switches per second Time/second InstanceId
          CpuInterruptSecond CPU Interrupts per second Time InstanceId
          vDiskReadBytesPerSecond Disk IO read throughput per second Byte/second InstanceId
          vDiskReadOpCountPerSecond Disk IO read operations per second Time InstanceId
          vDiskWriteBytesPerSecond Disk IO write throughput per second Byte/second InstanceId
          vDiskWriteOpCountPerSecond Disk IO write operations per second Time InstanceId
          DiskCFreeBytes Free space on Disk C Bytes InstanceId Exclusive to Windows server
          DiskCTotalBytes Total space on Disk C Bytes InstanceId Exclusive to Windows server
          DiskCUsedBytes Used space on Disk C Bytes InstanceId Exclusive to Windows server
          DiskCUsedPercent Disk C space utilization % InstanceId Exclusive to Windows server
          DiskFreeBytes Total free disk space on server Bytes InstanceId
          DiskFreeInodes Total free inodes on server Item InstanceId Exclusive to Linux server
          DiskInodesUsedPercent Total utilization of inodes on server % InstanceId Exclusive to Linux server
          DiskTotalBytes Total disk space on server Bytes InstanceId
          DiskTotalInodes Total inodes on server Item InstanceId Exclusive to Linux server
          DiskUsedBytes Total server disk utilization Bytes InstanceId
          DiskUsedInodes Total used inodes on server Item InstanceId Exclusive to Linux server
          DiskUsedPercent Server disk utilization % InstanceId
          RootUsedBytes Root disk space usage Bytes InstanceId Exclusive to Linux server
          RootUsedPercent Root disk space utilization % InstanceId Exclusive to Linux server
          HomeUsedBytes HOME disk space usage Bytes InstanceId Exclusive to Linux server
          HomeUsedPercent HOME disk space utilization % InstanceId Exclusive to Linux server
          MemAvailableBytes Available memory usage Bytes InstanceId Exclusive to Windows server
          MemBufferBytes Block device I/O memory buffer usage Bytes InstanceId Exclusive to Linux server
          MemCacheBytes File system memory cache value Bytes InstanceId
          MemFreeBytes Free memory Bytes InstanceId
          MemTotalBytes Total memory Bytes InstanceId
          MemUsedBytes Memory usage Bytes InstanceId
          MemUsedPercent Memory usage % InstanceId
          SwapFreeBytes Idle swap partition Bytes InstanceId Exclusive to Linux server
          SwapTotalBytes Total swap partition Bytes InstanceId Exclusive to Linux server
          SwapUsedBytes Swap partition usage Bytes InstanceId Exclusive to Linux server
          TcpCurrentEstab Established TCP connections Item InstanceId
          TcpInSegs TCP packets received Item InstanceId 1. Definition for Linux server: the rate of incoming TCP packets per second is calculated by reading the "InSegs" field in TCP within the /proc/net/snmp file, then computing the periodic difference divided by the time interval. 2. Definition for Windows server: the rate of incoming TCP packets per second is obtained via the Windows WMI.Win32_PerfFormattedData_Tcpip_TCPv4() interface.
          TcpLossSegs TCP error packets Item InstanceId 1. Exclusive to Linux servers. 2. Definition for Linux servers: the rate of incoming error packets per second is calculated by reading the "InErrs" field in TCP within the /proc/net/snmp file, then computing the periodic difference divided by the time interval.
          TcpOutSegs TCP packets sent Item InstanceId 1. Definition for Linux servers: the rate of outgoing TCP packets per second is calculated by reading the "OutSegs" field in TCP within the /proc/net/snmp file, then computing the periodic difference divided by the time interval. 2. Definition for Windows servers: the rate of outgoing TCP packets per second is obtained via the Windows WMI.Win32_PerfFormattedData_Tcpip_TCPv4() interface.
          TcpRetranSegs TCP retransmission count Time InstanceId 1. Exclusive to Windows servers. 2. Definition for Windows servers: the rate of retransmitted TCP packets per second is obtained via the Windows WMI.Win32_PerfFormattedData_Tcpip_TCPv4() interface.
          vNicInBytes Network interface card ingress traffic Bytes InstanceId
          vNicOutBytes Network interface card egress traffic Bytes InstanceId Description: The total egress traffic of the network interface card accumulated within the collection period, typically one minute.
          VNicInPPS Network interface card inbound packet rate pps InstanceId
          VNicOutPPS Network interface card transmit packet rate pps InstanceId
          VNicInBPS Network interface card inbound bandwidth bps InstanceId
          VNicOutBPS Network interface card outbound bandwidth bps InstanceId
          WebInBytes Ingress traffic to primary IP address from Internet Bytes InstanceId
          WebOutBytes Egress traffic from primary IP address to Internet Bytes InstanceId
          WebInBitsPerSecond Inbound bandwidth to primary IP address from Internet bps InstanceId
          WebOutBitsPerSecond Outbound bandwidth from primary IP address to Internet bps InstanceId
          WebInPPS Inbound packet rate to primary IP address from Internet pps InstanceId
          WebOutPPS Outbound packet rate from primary IP address to Internet pps InstanceId
          GpuError GPU card error message InstanceId Exclusive to GPU-equipped models
          GpuStatus GPU card overall status InstanceId Exclusive to GPU-equipped models
          GpuMaxEccErrorsIndex GPU ID with maximum ECC errors InstanceId Exclusive to GPU-equipped models
          GpuAllEccErrors ECC errors of all GPU cards Item InstanceId Exclusive to GPU-equipped models
          GpuMaxTemperatureIndex GPU ID with maximum temperature InstanceId Exclusive to GPU-equipped models
          GpuMaxTemperature Maximum temperature across all GPU cards InstanceId Exclusive to GPU-equipped models
          GpuMaxMemoryUtilizationIndex GPU ID with Maximum memory utilization InstanceId Exclusive to GPU-equipped models
          GpuMaxMemoryUtilization Maximum memory utilization across all GPU cards % InstanceId Exclusive to GPU-equipped models
          GpuMaxGpuUtilizationIndex GPU ID with maximum GPU utilization InstanceId Exclusive to GPU-equipped models
          GpuMaxGpuUtilization Maximum GPU utilization across all GPUs % InstanceId Exclusive to GPU-equipped models
          GpuAvgMemoryUtilizationForall Average memory utilization of all GPUs % InstanceId Exclusive to GPU-equipped models
          GpuAvgGpuUtilizationForall Average GPU utilization across all GPUs % InstanceId Exclusive to GPU-equipped models
          GPU{serial number}Error Error message for GPU card {serial number} InstanceId Exclusive to GPU-equipped models, {serial number} substituted with digits
          GPU{serial number}Status GPU card{serial number}status InstanceId Exclusive to GPU-equipped models, {serial number} substituted with digits
          GPU{serial number}UtilizationMemory GPU card{serial number}memory utilization % InstanceId Exclusive to GPU-equipped models, {serial number} substituted with digits
          GPU{serial number}EccErrors GPU card{serial number} ECC errors Item InstanceId Exclusive to GPU-equipped models, {serial number} substituted with digits
          GPU{serial number}Temperature GPU card{serial number} temperature InstanceId Exclusive to GPU-equipped models, {serial number} substituted with digits
          GPU{serial number}MemoryFree GPU card{serial number}free memory Bytes InstanceId Exclusive to GPU-equipped models, {serial number} substituted with digits
          GPU{serial number}MemoryUsed GPU card{serial number}memory usage Bytes InstanceId Exclusive to GPU-equipped models, {serial number} substituted with digits
          GPU{serial number}MemoryTotal GPU card{serial number}total memory Bytes InstanceId Exclusive to GPU-equipped models, {serial number} substituted with digits
          GPU{serial number}UtilizationGpu GPU card{serial number}GPU utilization % InstanceId Exclusive to GPU-equipped models, {serial number} substituted with digits
          CPUUsagePercent CPU usage % InstanceId Available on both Linux and Windows servers
          MemAlreadyUsedBytes Memory usage Bytes InstanceId Exclusive to Linux server, read/proc/meminfo, MemTotal - MemFree
          MemUserUsedBytes Actual user memory usage Bytes InstanceId Exclusive to Linux server, read/proc/meminfo, MemTotal - MemFree - Buffers - Cached - SReclaimable
          MemAvailableBytes Available memory usage Bytes InstanceId Exclusive to Linux server
          MemAvailablePercent Memory availability % InstanceId Exclusive to Linux server
          DiskXReadBytesPerSecond Disk read bandwidth Bytes/s InstanceId,disk Single VFIO local disk
          DiskXWriteBytesPerSecond Disk write bandwidth Bytes/s InstanceId,disk Single VFIO local disk
          DiskXReadOpCountPerSecond Disk read IOPS Time InstanceId,disk Single VFIO local disk
          DiskXWriteOpCountPerSecond Disk write IOPS Time InstanceId,disk Single VFIO local disk
          DiskXUsedBytes Single disk space usage Bytes InstanceId,disk Exclusive to Linux
          DiskXUsedPercent Single disk space utilization % InstanceId,disk Exclusive to Linux
          Disk[X]UsedByte Disk space usage Bytes InstanceId,disk Exclusive to Windows
          Disk[X]UsedPercent Disk space usage rate % InstanceId,disk Exclusive to Windows
          RdmaXmitPps RDMA interface transmit packet rate pps InstanceId,ip Exclusive to Linux
          RdmaRcvPps RDMA interface inbound packet rate pps InstanceId,ip Exclusive to Linux
          RdmaRcvBps RDMA interface inbound bandwidth bps InstanceId,ip Exclusive to Linux
          RdmaXmitBps RDMA interface outbound bandwidth bps InstanceId,ip Exclusive to Linux
          RdmaXmitDiscardsPps RDMA interface packet drop rate pps InstanceId,ip Exclusive to Linux
          RdmaLinkUp RDMA interface Up status - InstanceId,ip Exclusive to Linux
          RdmaSendCNP RDMA NIC sent CNP count Item/second InstanceId,ip Exclusive to Linux
          RdmaHandleCNP RDMA NIC processed CNP count Item/second InstanceId,ip Exclusive to Linux
          RdmaMarkedECN RDMA NIC marked ECN count Item/second InstanceId,ip Exclusive to Linux
          RdmaRcvPFC RDMA NIC received PFC count Item/second InstanceId,ip Exclusive to Linux
          RdmdXmitPFC RDMA NIC sent PFC count Item/second InstanceId,ip Exclusive to Linux
          RdmaACKTimeout RDMA NIC ACK timeout count Item/second InstanceId,ip Exclusive to Linux
          RDMAOutOfSequencePacket RDMA NIC out-of-order packet count Item/second InstanceId,ip Exclusive to Linux
          RdmaCRCError RDMA NIC CRC error count Item/second InstanceId,ip Exclusive to Linux
          GpuXUtilizationGpu GPU card utilization % InstanceId,gpu Exclusive to Linux
          GpuXStatus GPU card status - InstanceId,gpu Exclusive to Linux
          GpuXError GPU card error message - InstanceId,gpu Exclusive to Linux
          GpuXUtilizationMemory GPU card memory utilization % InstanceId,gpu Exclusive to Linux
          GpuXMemoryTotal Total GPU card memory Bytes InstanceId,gpu Exclusive to Linux
          GpuXMemoryFree Free GPU card memory Bytes InstanceId,gpu Exclusive to Linux
          GpuXMemoryUsed GPU card memory usage Bytes InstanceId,gpu Exclusive to Linux
          GpuXTemperature GPU temperature InstanceId,gpu Exclusive to Linux
          GpuXEccErrors ECC errors of GPU cards Item InstanceId,gpu Exclusive to Linux
          DCGM_GPU_TEMP GPU operating temperature InstanceId,gpu Exclusive to Linux
          DCGM_MEM_TEMP GPU memory temperature InstanceId,gpu Exclusive to Linux
          DCGM_FAN_SPEED_PERCENT GPU fan speed proportion % InstanceId,gpu Exclusive to Linux
          DCGM_POWER_USAGE GPU power W InstanceId,gpu Exclusive to Linux
          DCGM_GPU_PERF GPU performance state - InstanceId,gpu Exclusive to Linux
          DCGM_FI_DEV_TOTAL_ENERGY_CONSUMPTION Total GPU energy consumption since startup J InstanceId,gpu Exclusive to Linux
          DCGM_GPU_UTILIZATION GPU utilization % InstanceId,gpu Exclusive to Linux
          DCGM_ENC_UTILIZATION GPU encoder utilization % InstanceId,gpu Exclusive to Linux
          DCGM_DEC_UTILIZATION GPU decoder utilization % InstanceId,gpu Exclusive to Linux
          DCGM_MEM_COPY_UTILIZATION GPU memory copy utilization % InstanceId,gpu Exclusive to Linux
          DCGM_FB_FREE GPU frame buffer remaining MiB InstanceId,gpu Exclusive to Linux
          DCGM_FB_USED GPU frame buffer usage MiB InstanceId,gpu Exclusive to Linux
          DCGM_PROF_GR_ENGINE_ACTIVE GPU Graphics or Compute engine active time ratio % InstanceId,gpu Exclusive to Linux
          DCGM_PROF_SM_ACTIVE GPU SM active time ratio % InstanceId,gpu Exclusive to Linux
          DCGM_PROF_SM_OCCUPANCY GPU thread occupancy ratio on SM % InstanceId,gpu Exclusive to Linux
          DCGM_PROF_PIPE_TENSOR_ACTIVE GPU Tensor Pipe active cycle ratio % InstanceId,gpu Exclusive to Linux
          DCGM_PROF_PIPE_FP64_ACTIVE GPU FP64 pipe active cycle ratio % InstanceId,gpu Exclusive to Linux
          DCGM_PROF_PIPE_FP32_ACTIVE GPU FP32 pipe active cycle ratio % InstanceId,gpu Exclusive to Linux
          DCGM_PIPE_FP16_ACTIVE GPU FP16 pipe active cycle ratio % InstanceId,gpu Exclusive to Linux
          DCGM_PROF_DRAM_ACTIVE GPU memory bandwidth utilization % InstanceId,gpu Exclusive to Linux
          PROF_NVLINK_TX_BYTES NVLink data transfer rate Bytes InstanceId,gpu Exclusive to Linux
          PROF_NVLINK_RX_BYTES Nvlink data receive rate Bytes InstanceId,gpu Exclusive to Linux
          DCGM_FI_DEV_NVLINK_CRC_FLIT_ERROR_COUNT_TOTAL Total NVLink flow control CRC errors Item InstanceId,gpu Exclusive to Linux
          DCGM_FI_DEV_NVLINK_CRC_DATA_ERROR_COUNT_TOTAL Total number of NVLink data CRC errors. Item InstanceId,gpu Exclusive to Linux
          DCGM_FI_DEV_NVLINK_REPLAY_ERROR_COUNT_TOTAL Total NVLink retries Item InstanceId,gpu Exclusive to Linux
          DCGM_FI_DEV_NVLINK_RECOVERY_ERROR_COUNT_TOTAL Total NVLink recovery errors Item InstanceId,gpu Exclusive to Linux
          DCGM_FI_DEV_NVLINK_BANDWIDTH_TOTAL Total NVLink bandwidth counters Item InstanceId,gpu Exclusive to Linux
          PROF_PCIE_TX_BYTES GPU PCIe bus data transfer rate Bytes InstanceId,gpu Exclusive to Linux
          PROF_PCIE_RX_BYTES GPU PCIe bus data receive rate Bytes InstanceId,gpu Exclusive to Linux
          DCGM_PCIE_REPLAY_COUNTER GPU PCIe retry total - InstanceId,gpu Exclusive to Linux
          DCGM_SM_CLOCK GPU sm clock frequency HZ InstanceId,gpu Exclusive to Linux
          DCGM_MEMORY_CLOCK GPU memory clock frequency HZ InstanceId,gpu Exclusive to Linux
          DCGM_APP_SM_CLOCK GPU SM application clock frequency HZ InstanceId,gpu Exclusive to Linux
          DCGM_APP_MEMORY_CLOCK GPU memory application clock frequency HZ InstanceId,gpu Exclusive to Linux
          DCGM_CLOCK_THROTTLE_REASONS Reasons for GPU clock slowdown - InstanceId,gpu Exclusive to Linux
          DCGM_ECC_SBE_VOL_TOTAL Total GPU single-bit volatile ECC errors Item InstanceId,gpu Exclusive to Linux
          DCGM_ECC_DBE_VOL_TOTAL Total GPU double-bit volatile ECC errors Item InstanceId,gpu Exclusive to Linux
          DCGM_ECC_SBE_AGG_TOTAL Total GPU single-bit persistent ECC errors Item InstanceId,gpu Exclusive to Linux
          DCGM_ECC_DBE_AGG_TOTAL Total GPU double-bit persistent ECC errors Item InstanceId,gpu Exclusive to Linux
          DCGM_XID_ERRORS GPU XID error code - InstanceId,gpu Exclusive to Linux
          Previous
          Cloud native
          Next
          Elastic Baremetal Compute BBC