Performance Log Messages

Messages are logged that provide performance statistics for the agent. These messages include a performance area and statistic title for each value. This section describes the performance statistics found in the log messages.

Performance Area	Statistic Title	Description
CPU	Average Usage	The average CPU used by processes running in kernel mode and in user mode. Important The CPU average does not take into account the number of processors in the system. To obtain an accurate value, divide the average shown in the log message by the number of logical processors in the system. The times that each thread of the process ran in kernel or user mode is determined and totaled for each mode. These times can exceed the amount of real time elapsed if the process executes across multiple CPU cores. After these total times are calculated, the average CPU use is calculated for kernel mode and user mode. The difference between the current and previous kernel or user mode time is divided by elapsed time. This value is shown in the log message.
CPU	Total Usage	The number of seconds the agent ran in kernel mode and in user mode since the process started. The total time since the process started is given in parentheses.
IO	The counters in this section include all operations performed by all processes that have ever been associated with the job object. See Microsoft's IO_COUNTERS structure.
	BytesPerSec	The number of bytes per second read and written. These values represent the average over the number of seconds shown in parentheses in the message.
	Process IOPS	The number of read and write operations performed. These values represent the average over the number of seconds shown in parentheses in the message.
	Process TotalBytes	The total number of bytes read and written. The total time since the process started is given in parentheses.
KERN	The statistics in this section relate to Microsoft's PERFORMANCE_INFORMATION structure.
	HandleCount	The current number of open handles.
	Nonpaged	The memory currently in the nonpaged kernel pool, in pages.
	Paged	The memory currently in the paged kernel pool, in pages.
	ProcessCount	The current number of processes.
	ThreadCount	The current number of threads.
	Total	The sum of the memory currently in the paged and nonpaged kernel pools, in pages.
PROC	The statistics in this section relate to Microsoft's PROCESS_MEMORY_COUNTERS_EX structure.
	PageFaultCount	The number of page faults.
	PagefileUsage	The commit charge value, in bytes, for this process. Commit charge is the total amount of memory that the memory manager has committed for a running process.
	PeakPagefileUsage	The peak value, in bytes, of the commit charge during the lifetime of this process.
	PeakWorkingSetSize	The peak working set size, in bytes.
	Pool Usage: NonPaged	The current nonpaged pool usage, in bytes.
	Pool Usage: PeakNonPaged	The peak nonpaged pool usage, in bytes.
	Pool Usage: PeakPaged	The peak paged pool usage, in bytes.
	Pool Usage: QuotaPaged	The current paged pool usage, in bytes.
	WorkingSetSize	The current working set size, in bytes.
SYS	The statistics in this section relate to Microsoft's PERFORMANCE_INFORMATION structure.
	CommitLimit	The current maximum number of pages that can be committed by the system without extending the paging files.
	CommitPeak	The maximum number of pages that were simultaneously in a committed state since the last system reboot.
	CommitTotal	The number of pages currently committed by the system.
	PageSize	The size of a page, in bytes.
	PhysicalTotal	The amount of actual physical memory, in pages.
	SystemCache	The amount of system cache memory, in pages. This is the size of the standby list plus the system working set.