A complete list of all load test metrics (including .NET server metrics) can be found in the Load Test Metrics section.
Note that not all metrics are available on all operating systems.
CPU % (Processor Time for all processors) - the percentage of elapsed time that the processor spends executing non-Idle threads. This metric is the primary indicator of processor activity, and displays the average percentage of busy time observed during the sample interval. This metric is expected to increase proportionally to the load applied to the system.
Context switches/sec - the rate of switches from one thread to another. Thread switches can occur either inside of a single process or across processes. A thread switch can be caused either by one thread asking another for information, or by a thread being preempted by another, higher priority thread becoming ready to run. This metric is expected to increase proportionally to the load applied to the system.
Process Queue Length - the number of processes waiting to be scheduled to run. On windows, a sustained processor queue of less than 10 threads per processor is normally acceptable, dependent on the workload.
% Memory - The percentage of available memory that is in use. Large values of this metric suggest that the frequency of page swaps to disk will be high. For Windows servers, this is the percentage of virtual memory currently committed. For Linux servers, this is the percentage of physical memory in use (non free, cached or buffered).
Cache Memory Allocation - The amount of memory reserved by the Operating System for cache usage. Decreases in this value can be used as indicators that the Operating System requires memory for other purposes, which might not cause an immediate increase in memory usage.
Cache Memory Allocation Ratio - The percentage of physical memory reserved by the Operating System for cache usage. Decreases in this value can be used as indicators that the Operating System requires memory for other purposes, which might not cause an immediate increase in memory usage.
Page reads/sec - the rate at which the disk was read to resolve hard page faults. Hard page faults occur when a process references a page in virtual memory that is not in working set or elsewhere in physical memory, and must be retrieved from disk. This counter is a primary indicator of the kinds of faults that cause system-wide delays. It includes read operations to satisfy faults in the file system cache (usually requested by applications) and in non-cached mapped memory files. Large increases in this metric can degrade system performance. Increasing physical memory can alleviate the problem).
Page Writes/sec - the rate at which pages are written to disk to free up space in physical memory. Pages are written to disk only if they are changed while in physical memory, so they are likely to hold data, not code. Large increases in this metric can degrade system performance. Increasing physical memory can alleviate the problem.
% I/O Time Utilized - The percentage of time during the sample interval that the disk was executing I/O.
Service time: The average amount of time required for each I/O transfer.
Reads/sec - the rate of read operations on the disk. A plateau in this metric could indicate a performance bottleneck.
Writes/sec - the rate of write operations on the disk. A plateau in this metric could indicate a performance bottleneck.
Queue Length - The average number of write requests that were queued for the disk during the sample interval.
Packets Received/sec - the rate at which packets are received on the network interfaces. This metric is expected to increase proportionally to the applied load. A greater-than-linear increase could indicate less efficient operation of the network. A less-than-linear increase indicates a limitation of network and/or server capacity.
Packets Sent/sec - the rate at which packets are sent on the network interfaces. This metric is expected to increase proportionally to the applied load. A greater-than-linear increase could indicate less efficient operation of the network. A less-than-linear increase indicates a limitation of network and/or server capacity.
Bytes received/sec - the rate at which data is received on the network interfaces. This metric is expected to increase proportionally to the applied load. A greater-than-linear increase could indicate less efficient operation of the network. A less-than-linear increase indicates a limitation of network and/or server capacity.
Bytes sent/sec - the rate at which data is sent on the network interfaces. This metric is expected to increase proportionally to the applied load. A greater-than-linear increase could indicate less efficient operation of the network. A less-than-linear increase indicates a limitation of network and/or server capacity.
Packets Received Errors - the number of inbound packets that contained errors preventing them from being deliverable to a higher-layer protocol. These errors are considered a serious network degradation.
Packets Sent Errors - the number of outbound packets that contained errors preventing them from being deliverable to a higher-layer protocol. These errors are considered a serious network degradation.
Collisions/sec - the rate at which outgoing ethernet packets must be re-transmitted. When this metric exceeds 5% of packets sent/sec, this indicates a network problem or network capacity limit.
Connections Established - the number of TCP connections for which the current state is either ESTABLISHED or CLOSE-WAIT. This metric is expected to increase proportionally to the load applied to the system.
Connection Failures - the number of times TCP connections have gone from SYN-SENT or SYN-RCVD to CLOSED.
TCP Segments Retransmitted - the number of TCP segments which were previously transmitted, but had to be retransmitted again.
Developers or advanced users who need to see sources that these metrics are measured from should refer to the Locating Server Metric Counters section.