Host Metrics
VMware Aria
Operations for Networks
collects many
metrics for hosts, including CPU usage, datastore, memory, network, and disk
metrics.CPU Usage Metrics for
Hosts
These metrics provide information about the CPU use on the host.
Metric Name | Metric API Name | Description | Supported by Databus |
|---|---|---|---|
CPU Co-stop Time (%) | cpu.costop.delta.summation.millisecond | Percentage of time the host is ready, but unable to run due
to co-scheduling constraints. | |
CPU Ready Time (%) | cpu.ready.delta.summation.millisecond | Percentage of time the VM was waiting in line to use the CPU
on the host. | |
CPU Run Time (ms) | cpu.run.delta.summation.millisecond | Total time spent in run state by all VMs of
this host. | |
CPU Usage (MHz) | cpu.usagemhz.rate.average.megaHertz | Average CPU use in megahertz. | Yes |
CPU Usage Rate (%) | cpu.usage.rate.average.percent | Percentage of the CPU used out of its total capacity. | Yes |
CPU Wait Time (ms) | cpu.wait.delta.summation.millisecond | Total CPU time spent in wait state such as CPU idle, CPU swap
wait, and CPU I/O wait. | |
Uptime (s) | sys.uptime.absolute.latest.second | Total time elapsed since the last system startup. |
Network Metrics for Hosts
These metrics provide information about the network performance of the host.
Metric Name | Metric API Name | Description | Supported by Databus |
|---|---|---|---|
Dropped Rx Packets | net.droppedRx.delta.summation.number | Average ratio of the number of dropped Rx packets to the total
number of Rx packets. | Yes |
Dropped Tx Packets | net.droppedTx.delta.summation.number | Average ratio of the number of dropped Tx packets to the total
number of Tx packets. | Yes |
Max IO Latency | datastore.maxTotalLatency.absolute.latest.millisecond | Highest latency value across all datastores connected to the
host. | |
Max Network Traffic Rate (Kbps) | net.usage.rate.maximum.kiloBitsPerSecond | Maximum data transmission rate on the host over any five minute
interval in the selected duration. Maximum network rate
includes both transmitted and received data. | Yes |
Max Network Tx Rate (Kbps) | net.transmitted.rate.maximum.kiloBitsPerSecond | Maximum data transmission rate on the host over any five minute
interval in the selected duration. | Yes |
Max Packet Drop Ratio (%) | net.packetDrop.ratio.maximum.percent | Maximum ratio of total packets dropped to total number of packets
received over any five minute interval in the selected
duration. | Yes |
Max Packet Drops | net.droppedTotal.delta.maximum.number | Maximum number of dropped packets over any five minute interval
in the selected duration. | yes |
Multicast Rx Packets | net.multicastRx.delta.summation.number | Total number of multicast packets received across all pNICs
(physical network adapters) during the sampling interval. | Yes |
Multicast Tx Packets | net.multicastTx.delta.summation.number | Total number of multicast packets transmitted across all pNICs
during the sampling interval. | Yes |
Network Max Rx Rate (Kbps) | net.received.rate.maximum.kiloBitsPerSecond | Maximum network data receive rate on the host over any five
minute interval in the selected duration. | Yes |
Network Packet Count | net.packetsTotal.delta.summation.number | Total number of packets transmitted and received during the
interval across all pNICs. | Yes |
Network Packet Drop Percent (%) | net.packetDrop.ratio.average.percent | Ratio of total number of dropped packets to total number of
packets. | Yes |
Network Packet Drops | net.droppedTotal.delta.summation.number | Total number of packet drops on the host. | Yes |
Network Rx Rate (Kbps) | net.received.rate.average.kiloBitsPerSecond | Average rate at which data was received during the time
interval. | Yes |
Network Traffic Rate (Kbps) | net.usage.rate.average.kiloBitsPerSecond | Average rate of received and transmitted traffic during the time
interval. | Yes |
Network Tx Rate (Kbps) | net.transmitted.rate.average.kiloBitsPerSecond | Average rate at which data was transmitted during the time
interval. | Yes |
Outbound Expire Flow Drop Count | flow.fiveTupleExpireOutbound.delta.summation.number | Number of five tuples dropped due to the number of source IP
addresses of flows reaching the outbound limit. The outbound limit
for TCP flows is 1000 (allowed) and 50 (denied), and the
outbound limit of UDP flows is 500 (allowed) and 50
(denied). This
is a protection mechanism to make sure that the entities do not
overload the total flow limit. | |
Rx Broadcast Packets | net.broadcastRx.delta.summation.number | Number of broadcast packets received during a specified time
interval. | Yes |
Rx Broadcast Ratio | net.broadcastByRx.ratio.average.percent | Percent of broadcast packets received during a specified time
interval. | Yes |
Rx Packet Drop Percent (%) | net.packetDropRx.ratio.average.percent | Ratio of total number of dropped Rx packets to total number of Rx
packets. | Yes |
Rx Packets | net.packetsRx.delta.summation.number | Average number of received packets during the time
interval. | Yes |
Total Disk Latency (ms) | disk.totalLatency.absolute.average.millisecond | Average total latency (sum of kernel and device latency) for both
read and write commands. | |
Total Network Traffic | net.usage.delta.summation.kiloBytes | Sum of data transmitted and received for all the NIC instances of
the host. | Yes |
Tx Broadcast Packets | net.broadcastTx.delta.summation.number | Number of broadcast packets transmitted during a specified time
interval. | Yes |
Tx Broadcast Ratio | net.broadcastByTx.ratio.average.percent | Average percent of broadcast packets transmitted during a
specified time interval. | Yes |
Tx Packet Drop Percent (%) | net.packetDropTx.ratio.average.percent | Ratio of total number of dropped Tx packets to total number of Tx
packets. | Yes |
Tx Packets | net.packetsTx.delta.summation.number | Average number of transmitted packets during the time
interval. | Yes |
Unicast Rx Packets | net.unicastRx.delta.summation.number | Total number of unicast packets received by the host. | Yes |
Unicast Tx Packets | net.unicastTx.delta.summation.number | Total number of unicast packets transmitted by the host. | Yes |
Wrong Collector Flow Drop Count | flow.rawRecordReporterWrongCollector.delta.summation.number | Number of raw flow
records dropped as they were reported on a wrong type of
VMware Aria
Operations for Networks collector.Physical flows must
be reported to physical proxies. Other flows must be reported to
non-physical proxies. |
Datastore Metrics for
Hosts
These metrics provide information about the datastore use of the host.
Metric Name | Metric API Name | Description |
|---|---|---|
Read IOPS | datastore.numberReadAveraged.rate.average.number | Average number of read commands issued per second during the
collection interval. |
Read Latency (ms) | datastore.totalReadLatency.absolute.average.millisecond | Average latency of each read operation issued by the link through
which the host is connected to the datastore. |
Read Rate (Kbps) | datastore.read.rate.average.kiloBytesPerSecond | Average number of kilobytes read by the link through which the
host is connected to the datastore. |
RW IOPS | datastore.numberRWAveraged.rate.average.number | Average number of read and write commands issued per second
during the collection interval. |
RW Rate (Kbps) | datastore.usage.rate.average.kiloBytesPerSecond | The aggregated input rates and output rates of the link through
which the host is connected to the datastore. |
Write IOPS | datastore.numberWriteAveraged.rate.average.number | Average number of write commands issued per second during the
collection interval. |
Write Latency (ms) | datastore.totalWriteLatency.absolute.average.millisecond | Average latency of
each write operation issued by the link through which the host
is connected to the datastore. |
Write Rate (Kbps) | datastore.write.rate.average.kiloBytesPerSecond | Average number of kilobytes written by the link through which the
host is connected to the datastore. |
Memory Metrics for Hosts
These metrics provide information about memory use and allocation on the host.
Metric Name | Metric API Name | Description | Supported by Databus |
|---|---|---|---|
Memory Balloon | mem.vmmemctl.absolute.average.kiloBytes | The amount of memory currently claimed by the balloon driver. This metric
indicates that the host is taking memory from less needful VMs,
and allocating it to those VMs where there is an increased
memory requirement. | |
Memory State (0-3) | mem.state.absolute.latest.number | Overall state of the memory. The value is an integer between 0
(high) and 3 (low). | |
Memory Swap Used | mem.swapused.absolute.average.kiloBytes | The amount of swap memory currently in use. A large amount of
swap memory does not indicate a performance problem. Check the
swap rates to verify if the guest is actively in need of more
memory than that is available. | |
Memory Swap-In Rate (Kbps) | mem.swapinRate.rate.average.kiloBytesPerSecond | Rate at which the memory is swapped from disk to active memory
during the time interval. | |
Memory Swap-Out Rate (Kbps) | mem.swapoutRate.rate.average.kiloBytesPerSecond | Rate at which the memory is being swapped from active memory to
the disk during the time interval. | |
Memory Usage Rate (%) | mem.usage.absolute.average.percent | Percentage of memory used from the available machine
memory. | Yes |
Reserved Memory (KB) | mem.reservedCapacity.absolute.average.megaBytes | Amount of reserved memory by powered-on VMs and the memory used
for running VMware vSphere services on the host. | |
Overhead Memory (KB) | mem.overhead.absolute.average.kiloBytes | Memory overhead reported by the host. Memory overhead is
the total memory allocated to powered on VMs beyond its reserved
memory and the memory used for running vSphere services on the
host. | |
Consumed Memory (KB) | mem.consumed.absolute.average.kiloBytes | Total memory consumed by the VMs and VMware vSphere services in
the host. | |
Active Memory (KB) | mem.active.absolute.average.kiloBytes | Total memory actively used by the host. Total memory includes
memory used by VMs and VMware vSphere services running on the
host. The VMkernel estimates the total memory based on recently
touched memory pages. | Yes |
Disk Metrics for Hosts
These metrics provide information about disk use of the host.
Metric Name | Metric API Name | Description |
|---|---|---|
Average Read Latency (ms) | disk.deviceReadLatency.absolute.average.millisecond | Average latency of each read operation issued to complete a Small
Computer System Interface (SCSI) command from the physical
device. |
Average Write Latency (ms) | disk.deviceWriteLatency.absolute.average.millisecond | Average latency of each write operation issued to complete a SCSI
command from the physical device. |
Disk Bus Resets | disk.busResets.delta.summation.number | Number of SCSI-bus reset commands issued during the collection
interval. |
Disk Commands Aborted | disk.commandsAborted.delta.summation.number | Number of SCSI commands aborted during the collection
interval. |
Disk Device Latency (ms) | disk.deviceLatency.absolute.average.millisecond | Average amount of time taken to complete a SCSI command from the
physical device. |
Disk Kernel Latency (ms) | disk.kernelLatency.absolute.average.millisecond | Average amount of time spent by the VMkernel processing each SCSI
command. |
Disk Queue Depth | disk.queueDepth.absolute.latest.number | Maximum queue depth during the collection interval. |
Disk Queue Latency | disk.queueLatency.absolute.average.millisecond | Average amount of time spent in the VMkernel queue, per SCSI
command, during the collection interval. |
Disk Reads (KB) | disk.read.rate.average.kiloBytesPerSecond | Average number of reads through this link. |
Disk Writes (KB) | disk.write.rate.average.kiloBytesPerSecond | Average number of writes through this link. |
Read IOPS | disk.numberReadAveraged.rate.average.number | Average number of reads issued to complete a SCSI command through
this link. |
RW IOPS | disk.numberRWAveraged.rate.average.number | Average number of reads or writes issued to complete a SCSI
command through this link. |
Total Disk Latency (ms) | disk.totalLatency.absolute.average.millisecond | Average amount of time taken during the collection interval to
process a SCSI command issued by the guest operating system to the
VM. |
Write IOPS | disk.numberWriteAveraged.rate.average.number | Average number of writes issued to complete a SCSI command
through this link. |