TOCPREVNEXTINDEX

Lund Performance Solutions


SOS Workload Detail

The Workload Detail Screen

The Workload Detail screen displays detailed information about a specific workload. For information about workloads, refer to "Workload Groups".


Figure 28.1 SOS Workload Detail screen
To access the Workload Detail screen from any SOS display screen:
  • Type s from the SOS Enter command: prompt to view the Screen Selection Menu screen.
  • From the Screen Selection Menu screen, enter W (Workload Detail).
  • At the secondary prompt:
  • Press the Enter key to display the workload detail information for the given workload.
  • Or, select another workload from the list.
  • The Workload Detail screen will display. An example of the screen is shown in Figure 28.1.
  • Use the y command key to toggle between the normal and extended process displays.
  • Workload Detail Display Items

    WORKLOAD DETAIL

    The WORKLOAD DETAIL portion of the Workload Detail screen displays workload identification information. The data items are described in the next table.
    Table 28.1 SOS WORKLOAD DETAIL data items
    Data Item
    Description
    Workload
    The name of the workload.
    Proc Count
    The average number of processes that exists within the workload.

    CPU

    The CPU portion of the Workload Detail screen displays process and workload CPU information. The data items are described in Table 28.2.
    Table 28.2 SOS CPU data items
    Data Item
    Description
    CPU %
    The percentage of the current interval that the process was executing. This value is normalized for multiple processors—the sum of the CPU% values should not exceed 100.
    CPU ms
    The number of milliseconds the process was executing. The value in brackets, [ ], is a cumulative value, not an averaged cumulative value.
    User %
    The percentage of the process’ execution time spent in user mode. This includes real, nice, and negative nice time.
    Sys %
    The percentage of the process’ execution time spent in system/kernel mode. This includes memory and trap time.
    Trap %
    The percentage of the process’ execution time spent processing traps.

    MEMORY

    The MEMORY portion of the Workload Detail screen displays process and workload memory information. The data items are described in the following table.
    Table 28.3 SOS MEMORY data items
    Data Item
    Description
    RSS
    The resident set size, which is equal to the amount of RAM the process is using. This value does not include shared memory.
    VSS
    The amount of virtual memory the process has reserved, which is equal to the size of the process’ core image including text, data, and stack.
    State
    The process state with respect to memory (not applicable to workloads). The possible values are:
  • Dead (the process is dead)
  • LOAD (the process is loaded in memory)
  • DEACT (the process is deactivated)
  • Min /s
    The number of minor page faults experienced by the process, per second.
    Maj /s
    The number of major page faults experienced by the process, per second.
    Swp /s
    The number of processes swapped out per second.

    RESP TIME

    The RESP TIME portion of the Workload Detail screen displays process and workload response time and transaction information. The data items are described in Table 28.4.
    Table 28.4 SOS RESP TIME data items
    Data Item
    Description
    Tran /s
    The number of transactions per second.
    Resp T
    The average time (milliseconds) for a transaction to complete.

    DISK

    The DISK portion of the Workload Detail screen displays process and workload disk statistics. The data items are described in Table 28.5.
    Table 28.5 SOS DISK data items
    Data Item
    Description
    Phyio in
    The number of physical disk reads.
    Phyio out
    The number of physical disk writes.
    Phyio in/s
    The number of physical disk reads per second.
    Phyio out/s
    The number of physical disk writes per second.
    IO%
    The percentage of I/Os performed on the disk, system-wide.

    WAIT STATES

    The WAIT STATES portion of the Workload Detail screen displays various process and workload wait state information. The data items are described in the following table.
    Table 28.6 SOS WAIT STATES data items
    Data Item
    Description
    JOB
    The percentage of a process’ time spent in the JOB state, waiting for job control or tracing signals.
    CPU
    The percentage of a process’ time spent in the CPU state.
    OTHR
    The percentage of a process’ time spent in the OTHR state, waiting on all other resources and events.
    PRI
    The percentage of a process’ time spent in the PRI state, waiting for the CPU.
    TFLT
    The percentage of a process’ time spent in the TFLT state, waiting on text page faults.
    DFLT
    The percentage of a process’ time spent in the DFLT state, waiting on data page faults.
    KFLT
    The percentage of a process’ time spent in the KFLT state, waiting on kernel page faults.
    ULCK
    The percentage of a process’ time spent in the UFLT state, waiting on user locks.

    PROCESS SUMMARY

    The PROCESS SUMMARY portion of the Workload Detail screen displays general information about the processes on the system. By default, any process that was active in the interval will be displayed. The information display can be reconfigured using the commands described in "Process display option Commands". The data items are described in Table 28.7.
    Table 28.7 SOS WORKLOAD DETAIL data items
    Data Item
    Description
    PID
    The product identification number of the process.
    Name
    The name of the process—the command used to create the process.
    User Name
    The name of the user that owns the process.
    Tty
    The device file associated with the terminal to which the process is attached. If no terminal is associated with the process, three dashes (---) will display.
    CPU%
    The percentage of system-wide CPU time that was used by the process. This value is normalized for multiple processors—the sum of the CPU% values should not exceed 100.
    Nice
    The nice value. A value of R indicates the process has a real time priority—the nice value is not used. For information about the nice utility, see the Unix man page, "nice."
    Pri
    The priority of the process. The value depends on the scheduling policy.
    RSS/Size
    The resident set size—the amount of RAM used by the process. The Size value represents the size in kilobytes of the core image of the process. This includes text, data, and stack space. In other words, the amount of swap or virtual memory the process has reserved.
    #Rd
    The number of physical reads performed by the process.
    #Wr
    The number of physical writes performed by the process.
    State
    The State data item shows the current state of the process. Possible values are:
  • CPU (the process is waiting for CPU)
  • RUN (the process is running on the CPU)
  • SLEEP (the process is waiting for a resource or event)
  • STOP (the process is suspended by job control because it is being traced)
  • ZOMB (the process is dead, but still contains a process table entry)

  • Lund Performance Solutions
    www.lund.com
    Voice: (541) 812-7600
    Fax: (541) 81207611
    info@lund.com
    TOCPREVNEXTINDEX