Contents
In addition to system performance metrics from the ESX platform, Uptime Infrastructure Monitor also collects workload data for each instance running on an ESX server. These metrics show the consumption of the various physical resources by each ESX instance.
Uptime Infrastructure Monitor collects the following workload data (as documented in the VMware specifications) from an instance running on an ESX system:
| Metric | Explanation | Group/VMware Performance Counter |
| CPU | The percentage of CPU time that is being used by a VMware instance. This is a percentage of the available maximum amount of CPU time. | cpu/usagemhz |
| Memory | The amount of physical memory, in kilobytes, that is being used by a VMware instance. | mem/active |
| Network IO | The amount of the network I/O capacity, in kilobits per second, that is being used by a VMware instance. | net/usage |
| Disk IO | The amount of the disk I/O capacity, in kilobytes per second, that is being used by a VMware instance. | disk/usage |
| % Ready | The amount of time that one or more instances running on an ESX v3 server is ready to run, but cannot run because it cannot access the processor on the ESX v3 server. | cpu/ready |
| % Used | The percentage of CPU time that an instance running on an ESX v3 server is using. | cpu/used |
The following metrics are gathered by the ESX Workload service monitor, which is deployed to monitor instances running on an existing ESX system:
| Metric | Explanation |
| CPU Usage | The amount of processor power, measured in megahertz (MHz), that the instances on the ESX server are consuming. |
| Network Bandwidth Usage | The amount of network traffic in and out of the server, measured in megabits per second (Mbit/s). |
| Disk Usage | The amount of data being written to the server’s hard disk, measured in kilobytes per second (kB/s). |
| Memory Usage | The amount of overall system memory, measured in megabytes (MB). |
| Percent Ready | The percentage of time that one or more instances running on an ESX server is ready to run, but cannot run because it cannot access the processor on the ESX server. |
| Percent Used | The percentage of CPU time that an instance running on an ESX server is using. |
The following metrics are gathered by the ESX Advanced Metrics service monitor, which is deployed to monitor instances running on an existing ESX system:
| Metric | Explanation | Guest or Host Metric |
| Percent Wait | The percentage of CPU time that an instance running on an ESX server spent in a wait state. | guest |
| Memory Balloon | The amount of memory, in KB, allocated by the virtual machine memory control driver (vmmemctl, installed with VMware Tools). It’s a VMware exclusive memory-management driver that controls ballooning. | guest |
| Memory Balloon Target | The target value, in KB, set by VMkernal for the virtual machine’s memory balloon size. In conjunction with the vmmemctl metric, this metric is used by VMkernel to inflate and deflate the balloon for a virtual machine. | guest |
| Memory Overhead | The amount of machine/host memory, in KB, allocated to a virtual machine beyond its reserved amount. | guest |
| Memory Swap In | The amount of data, in KB, that has been swapped in to machine memory from the swap file since the virtual machine was powered on. | guest |
| Memory Swap Out | The amount of data, in KB, that the VMkernel memory has written to the virtual machine's swap file from machine memory. Refers to VMkernel swapping and not to guest OS swapping. | guest |
| Memory Zero | The amount of guest physical memory, in KB, that contains only 0s, thus can be safely used by other virtual machines for their 0 pages memory requirement. | guest |
| Memory Swap Used | The average amount of memory, in KB, that is used by swap. Swap is the sum of memory swapped for all VMs powered on, and vSphere services on the host. | host |
| Memory Swap Target | The amount of memory, in KB, available for swapping, where the target size for a virtual machine swap file is calculated by the VMkernel. | guest |
| Disk Total Latency | The average amount of time, in milliseconds, taken during the collection interval to process a SCSI command issued by the Guest OS to the virtual machine. The sum of kernelCommandLatency and physical deviceCommandLatency. | host |
| Disk Kernel Latency | The average amount of time, in milliseconds, spent by VMkernel processing each SCSI command. | host |
| Disk Device Latency | The average amount of time, in milliseconds, taken to complete a SCSI command from the physical device. | host |
| Disk Queue Latency | The average amount of time, in milliseconds, spent in the VMkernel queue, per SCSI command, during the collection interval. | host |
| Disk Commands Aborted | The number of SCSI commands aborted during the collection interval. | host |
| Disk Commands Issued | The number of SCSI commands issued during the collection interval. | host |
| Disk Bus Resets | The number of SCSI-bus reset commands issued during the collection interval. | host |