Monitoring and Tuning Commands and Subroutines
Performance Reporting and Analysis Commands
Performance reporting and analysis commands give you information on the performance of one or more aspects of the system, or on one or more of the parameters that affect performance. The commands are as follows:
Command
Function
alstat
Reports alignment exceptions counts
atmstat
Shows Asynchronous Transfer Mode (ATM) adapter statistics
emstat
Reports emulation instruction counts
entstat
Shows ethernet device driver and device statistics
fddistat
Shows FDDI device driver and device statistics
filemon
Uses the trace facility to report on the I/O activity of physical volumes, logical volumes, individual files, and the Virtual Memory Manager
fileplace
Displays the physical or logical placement of the blocks that constitute a file within the physical or logical volume on which they reside
gprof
Reports the flow of control among the subroutines of a program and the amount of CPU time consumed by each subroutine
ifconfig
Configures or displays network interface parameters for a network using TCP/IP
iostat
Displays utilization data for:
Terminals
CPU
Disks
Adapters (AIX 5.1 and subsequent releases)
ipfilter
Extracts different operation headers from an ipreport output file and displays them in a table
ipreport
Generates a packet trace report from the specified packet trace file
iptrace
Provides interface-level packet tracing for Internet protocols
lockstat
Displays information about kernel lock contention (only available in AIX Version 4)
locktrace
Turns on lock tracing (only available in AIX 5.1 and subsequent releases)
lsattr
Displays attributes of the system that affect performance, such as:
Processor speed (AIX 5.1 and subsequent releases)
Size of the caches
Size of real memory
Maximum number of pages in the block I/O buffer cache
Maximum number of kilobytes of memory allowed for mbufs
High- and low-water marks for disk-I/O pacing
lsdev
Displays devices in the system and their characteristics
lslv
Displays information about a logical volume
lsps
Displays the characteristics of paging spaces
lspv
Displays information about a physical volume within a volume group
lsvg
Displays information about volume groups
netpmon
Uses the trace facility to report on network activity, including:
CPU consumption
Data rates
Response time
netstat
Displays a wide variety of configuration information and statistics on communications activity, such as:
Current status of the mbuf pool
Routing tables
Cumulative statistics on network activity
nfso
Displays (or changes) the values of NFS options
nfsstat
Displays statistics on Network File System (NFS) and Remote Procedure Call (RPC) server and client activity
no
Displays (or changes) the values of network options, such as:
Default send and receive socket buffer sizes
Maximum total amount of memory used in mbuf and cluster pools
pdt_config
Starts, stops, or changes the parameters of the Performance Diagnostic Tool
pdt_report
Generates a PDT report based on the current historical data
pprof
Reports CPU usage of all kernel threads over a period of time
prof
Displays object file profile data
ps
Displays statistics and status information about the processes in the system, such as:
Process ID
I/O activity
CPU utilization
sar
Displays statistics on operating-system activity, such as:
Directory accesses
Read and write system calls
Forks and execs
Paging activity
schedtune
Displays (or changes) the values of VMM memory-load-control parameters, the CPU-time-slice duration, and the paging-space-low retry interval
smitty
Displays (or changes) system-management parameters
svmon
Reports on the status of memory at system, process, and segment levels
tcpdump
Prints out packet headers
time, timex
Prints the elapsed and CPU time used by the execution of a command
topas
Reports selected local system statistics
tokstat
Shows Token-Ring device driver and device statistics
tprof
Uses the trace facility to report the CPU consumption of kernel services, library subroutines, application-program modules, and individual lines of source code in the application program
trace, trcrpt
Writes a file that records the exact sequence of activities within the system
traceroute
Prints the route that IP packets take to a network host
vmstat
Displays VMM data, such as:
Number of processes that are dispatchable or waiting
Page-frame free-list size
Page-fault activity
CPU utilization
vmtune
Displays (or changes) the page-replacement algorithm parameters for the Virtual Memory Manager (VMM)
The following tools allow you to change one or more performance-related aspects of the system.
Command
Function
bindprocessor
Binds a process to a processor
chdev
Changes the characteristics of a device
chlv
Changes only the characteristics of a logical volume
chps
Changes attributes of a paging space
fdpr
Optimizes executable files for a specific workload
ifconfig
Configures or displays network interface parameters for a network using TCP/IP
migratepv
Moves allocated physical partitions from one physical volume to one or more other physical volumes
mkps
Adds an additional paging space to the system
nfso
Changes (or displays) the values of NFS options
nice
Executes a command at a specified priority
no
Changes (or displays) the values of network options
renice
Changes the priority of running processes
reorgvg
Reorganizes elements of a volume group
rmss
Temporarily reduces the effective RAM size of a system to assess the probable performance of a workload on a smaller machine or to ascertain the memory requirement of one element of a workload
schedtune
Changes (or displays) the values of VMM memory load control parameters, the CPU-time-slice duration, and the paging-space-low retry interval
smitty
Changes (or displays) system-management parameters
vmtune
Changes (or displays) the Virtual Memory Manager page-replacement algorithm parameters
The following subroutines can be used in monitoring and tuning performance:
bindprocessor()
Binds kernel threads to a processor
getpri()
Determines the scheduling priority of a running process
getpriority()
Determines the nice value of a running process
getrusage()
Retrieves information about the use of system resources
nice()
Increments the nice value of the current process
psdanger()
Retrieves information about paging space use
setpri()
Changes the priority of a running process to a fixed priority
setpriority()
Sets the nice value of a running process
