CPU and Memory Central Display

This display is available for the following platforms: LINUX NONSTOP UNIX WINDOWS

CPU and Memory data is provided through two different Displays, with each one based on the Prognosis Server platform.

MS Windows

The Windows CPU and Memory Overview Display shows critical system resource utilization for a single Windows server. It makes use of a number of Records including, NtProcess (NTJOBS), NtCPU (NTCPU), NtMemory (NTMEMORY) and NtSystem (NTSYSTEM).

This Display can be accessed from the Windows Server Overview Display by clicking on the CPU and Memory link in the 'Server Navigation' window.

This Display includes a table showing the '20 Busiest Processes' and two graphs showing 'Average CPU Busy' and 'Memory Utilization'. In addition, data is shown for Physical and Virtual Memory.

20 Busiest Processes

The 20 Busiest Processes table shows the following data.

PID

The process ID is the unique identifier of this process. PIDs are reused, so they only identify a process for the lifetime of that process. Drilling-down on a PID will open the Process Details Display that shows more detailed information.

Name

The name of the process. This is not a unique identifier as there may be many instances of the same process running at the same time. Drilling-down on a Name will open the Processes by Name Display.

Busy%

Busy Percent is the percentage of elapsed time that all of the threads of this process used in the processor to execute instructions. An instruction is the basic unit of execution in a computer, a thread is an object that executes instructions and a process is the object created when a program is run. Code executed to handle certain hardware interrupts or trap conditions may be counted for this process.

Faults/s

Page Faults per second is the rate of Page Faults by the threads executing in this process. A page fault occurs when a thread refers to a virtual memory page that is not in its working set in main memory. This will not cause the page to be fetched from disk if it is on the standby list and hence already in main memory, or if it is in use by another process with whom the page is shared. Drilling-down on a fault will open the Page Fault Details for Process Display.

Private

Private Bytes is the current number of megabytes this process has allocated that cannot be shared with other processes.

Virtual

Virtual MegaBytes is the current size in megabytes of the virtual address space the process is using. The use of virtual address space does not necessarily imply corresponding use of either disk or main memory pages. Virtual space is however finite, and by using too much, the process may limit its ability to load libraries.

At the top right of the Display, there are three links that are used to open Reports for the selected statistics.

Memory

Opens the Free Memory Last Hour report

CPU

Opens the CPU Busy % - Last Hour report

Processes

Opens the Server Health - Last Hour Display

At the bottom left of the Display, a number of links are provided to access further detailed information.

Processes

Clicking on the number of Processes will open the Processes Display where detailed information about all running Processes is shown.

Page Faults/sec

Clicking on the number of page faults will open the Page Faulting Processes Display which shows a summary of page fault details.

Handles

Clicking on the number of handles will open the Handles Central Display which shows critical system resource utilization.

Ports

Opens the Ports Open by All Processes Display.

Registry

Opens the Registry Operations Central Display which shows the top 10 busiest registry keys and the processes accessing them.

File Operations

Opens the File Operations Central Display which shows the top 10 busiest disk files and the processes that are accessing them.

UNIX/Linux

The CPU and Memory Central Display shows critical system resource utilization for a single UNIX server. It makes use of a number of Records including, UnixCpu (UXCPU), UnixProcess (UXPROC), and UnixSwapDevice (UXSWAP).

This Display can be accessed from the UNIX Server Overview Display by clicking on the CPU and Memory link in the 'Server Navigation' window.

The CPU and Memory Central Display consists of a set of Key Performance Indicators as well as 3 main windows, each containing a data chart or table; these are 'CPU Busy Last 5 Min', 'Free Swap Space MB' and 'Busiest 10 Processes'. The function of each of these is as follows:

Key Performance Indicators

The Windows at the top of the Display, under the link bar, show 'KPIs' or 'Key Performance Indicators'. These provide useful figures that reflect the CPU and Memory state of the UNIX machine.

Average and Max CPU Busy indicators show the busiest CPU, and the average CPU Busy % for all CPUs on this UNIX machine.

Min/Avg/Total Swap Space Free is useful for helping you to spot the smallest Free Swap Space, comparing that to the Average Free Swap Space, and then seeing the total Swap Space all at a glance.

CPU Busy Last 5 Min

This provides a graph of the minimum, maximum and average CPU busy percentages for the UNIX machine in the last five minutes. It provides a picture of how busy the machine has been, and can indicate if there are any spikes or drops in CPU usage in the last five minutes.

CPU Busy by Workload

This window shows CPU Busy by Workload Management groups. A Workload is a collection of processes that have been grouped together using rules defined in the Workload Configuration file. For details see the Workload Management for Linux/UNIX.

Busiest 10 Processes

This Window provides a table of the 10 busiest processes running on the UNIX machine, including all transient processes. It is sorted with the busiest processes at the top of the table.

The CPU Busy % field is color-coded for easy spotting of cpu-intensive or troublesome processes. If a process is using less than 30% of CPU, this field is highlighted green. If it is using more than 30% but less than 50%, it will be light blue. If it is using between 50% and 90% it will be orange, and a red color means that process is using more than 90% of CPU.

The system can show the transient processes too (see the 'Transient Processes' link at the top right of the Display), any transient processes displayed will have a blue-highlighted 'Start Time' field. This is useful for spotting transients, which may be using a large percentage of CPU collectively yet are hard to spot at times individually.

Drill-downs

The 'Busiest 10 Processes' window contains a number of Drill-downs that will open further Displays containing a range of useful information. Depending on which field is drilled-down on, the data available includes everything from a very detailed view about a specific process to an overview of processes running for a specific user.

Links

At the top of the CPU and Memory Central Display, a number of links are provided to open further Displays.

Workload Central

Opens the Workload Central Display. This provides details of Workload groups that have been created on the server together with CPU Usage and Page Fault data. For details see the Workload Management for Linux/UNIX.

Transient Processes

Opens the Transient Processes Central Display, which shows transient processes by executable name, by user, and compares transient CPU usage to total CPU usage on this UNIX machine.

CPU Details

Opens a Display showing CPU specific details for all CPUs on this UNIX machine.

Mem Details

Takes you to the UNIX Memory and Buffers Central Display.

Processes

Opens a Display showing process details including transient processes on this UNIX machine.

Ports Open

Opens the Ports Central Display.

Files Open

Opens the Files Open Central Display.

Kernel

Opens the Unix Kernel Table Details Display.

Queues

Opens a Display showing Queue Size statistics on this UNIX machine.

File Sys

Opens the File System Access Statistics Display.

HPE NonStop

The NonStop - CPU and Memory Central Display shows critical system resource utilization for a single HPE NonStop server. Making use of the NonStopCpu (CPU) and NonStopJob (JOBS) Records it provides an overview of CPU usage detailing the busiest processes and available memory. The benefits of this data include being able to:

  • Easily monitor and eliminate unnecessary CPU/memory usage.

  • See if anything is writing to disk and hence increasing CPU usage.

  • Observe how much impact an individual user is having on CPU usage.

  • Pinpointing troublesome processes and being able to drill-down to the details of a particular process for further investigation or monitoring.

This Display can be accessed from the NonStop Server Overview Display by clicking on the CPU And Memory link in the 'Server Navigation' window.

The CPU and Memory Central Display consists of a set of Key Performance Indicators as well as three main Windows each containing a data chart or table; these are 'CPU Busy %', 'CPU Memory Free Pages' and 'Busiest Processes'. The function of each of these is as follows:

Key Performance Indicators

The Windows at the top of the Display under the link bar, show 'KPIs' or 'Key Performance Indicators'. These provide useful figures that reflect the CPU and Memory state of the monitored HPE NonStop machine.

Average and Busiest CPU Busy indicators show the busiest CPU, and the average CPU Busy % for all CPUs on this machine. The Average Free Pages provides the average free memory pages over all CPUs on the machine, whereas Min Free Mem shows which CPU has the least free memory pages.

CPU Busy % - Min/Max/Avg

This line graph provides details of the minimum, maximum and average CPU busy percentages for the machine over the previous three minutes. It provides a picture of how busy the machine has been, and can indicate if there are any spikes or drops in CPU usage in the last three minutes.

CPU Memory Free Pages

This line graph shows the Minimum/Average/Maximum free memory pages over all CPUs on this machine and this can be used to help spot memory problems. For example, if the system is healthy and one CPU is running low on free memory, the minimum CPU memory free pages will be significantly lower than the average therefore indicating a possible problem. The graph will show whether this is temporary or whether this has been the trend for the past three minutes.

Busiest Processes

This Window provides a table of all processes running on the machine that are using more than 1% of CPU. It is sorted with the busiest processes at the top of the table.

Links

At the top of the CPU and Memory Central Display a number of links are provided to open further Displays:

ServerNet Central

Opens the NonStop - ServerNet Central Display.

OSS Performance

Opens the OSS Performance Monitor Display.

CPU Details

Takes you to the NonStop - CPU Details Display, which shows Usage, I/O and Memory information for all CPUs on this HPE NonStop machine.

Memory Details

Opens the Nonstop - Memory Details Display showing memory information for all CPUs on this HPE NonStop machine.

Provide feedback on this article