Performance Management Guide--MEM

netyu

Performance Management Guide
Determining How Much Memory Is Being Used
Several performance tools provide reports of memory usage. The reports of most interest are from the vmstat, ps, and svmon commands.
The vmstat Command (Memory)
The vmstat command summarizes the total active virtual memory used by all of the processes in the system, as well as the number of real-memory page frames on the free list. Active virtual memory is defined as the number of virtual-memory working segment pages that have actually been touched (for a definition see
Late Page Space Allocation
). This number can be larger than the number of real page frames in the machine, because some of the active virtual-memory pages may have been written out to paging space.
When determining if a system might be short on memory or if some memory tuning needs to be done, run the vmstat command over a set interval and examine the pi and po columns on the resulting report. These columns indicate the number of paging space page-ins per second and the number of paging space page-outs per second. If the values are constantly non-zero, there might be a memory bottleneck. Having occasional non-zero values is not be a concern because paging is the main principle of virtual memory.
# vmstat 2 10
kthr     memory             page              faults        cpu
----- ----------- ------------------------ ------------ -----------
r  b   avm   fre  re  pi  po  fr   sr  cy  in   sy  cs us sy id wa
1  3 113726   124   0  14   6 151  600   0 521 5533 816 23 13  7 57
0  3 113643   346   0   2  14 208  690   0 585 2201 866 16  9  2 73
0  3 113659   135   0   2   2 108  323   0 516 1563 797 25  7  2 66
0  2 113661   122   0   3   2 120  375   0 527 1622 871 13  7  2 79
0  3 113662   128   0  10   3 134  432   0 644 1434 948 22  7  4 67
1  5 113858   238   0  35   1 146  422   0 599 5103 903 40 16  0 44
0  3 113969   127   0   5  10 153  529   0 565 2006 823 19  8  3 70
0  3 113983   125   0  33   5 153  424   0 559 2165 921 25  8  4 63
0  3 113682   121   0  20   9 154  470   0 608 1569 1007 15  8  0 77
0  4 113701   124   0   3  29 228  635   0 674 1730 1086 18  9  0 73
Notice the high I/O wait in the output and also the number of threads on the blocked queue. Most likely, the I/O wait is due to the paging in/out from paging space.
To see if the system has performance problems with its VMM, examine the columns under memory and page:

• memory
  Provides information about the real and virtual memory.
  
  
  
  • avm
    The avm (Active Virtual Memory) column gives the average number of 4 K pages that are allocated to paging space. The avm value can be used to calculate the amount of paging space assigned to executing processes.
    Note: Starting with AIX 4.3.2, there is a slight change in reporting this value. See
    Looking at Paging Space and Virtual Memory
    for an explanation.
    The number in the avm field divided by 256 will yield the approximate number of megabytes (MB) allocated to paging space systemwide. Prior to AIX 4.3.2 the same information is reflected in the Percent Used column of the lsps -s command output or with the svmon -G command under the pg space inuse field.
    
  • fre
    The fre column shows the average number of free memory pages. A page is a 4 KB area of real memory. The system maintains a buffer of memory pages, called the free list, that will be readily accessible when the VMM needs space. The minimum number of pages that the VMM keeps on the free list is determined by the minfree parameter of the vmtune command (for details, see
    Tuning VMM Page Replacement with the vmtune Command
    ).
    When an application terminates, all of its working pages are immediately returned to the free list. Its persistent pages (files), however, remain in RAM and are not added back to the free list until they are stolen by the VMM for other programs. Persistent pages are also freed if the corresponding file is deleted.
    For this reason, the fre value may not indicate all the real memory that can be readily available for use by processes. If a page frame is needed, then persistent pages related to terminated applications are among the first to be handed over to another program.
    If the fre value is substantially above the maxfree value, it is unlikely that the system is thrashing. Thrashing means that the system is continuously paging in and out. However, if the system is experiencing thrashing, you can be assured that the fre value will be small.
    
  
  
• page
  Information about page faults and paging activity. These are averaged over the interval and given in units per second.
  
  
  
  • re
    Note: In AIX Version 4, reclaiming is no longer supported because the value it delivered by providing limited information about the performance of the system was outweighed by the negative affect on system performance due to the algorithm that kept track of reclaims.
    
  • pi
    The pi column details the number (rate) of pages paged in from paging space. Paging space is the part of virtual memory that resides on disk. It is used as an overflow when memory is over committed. Paging space consists of logical volumes dedicated to the storage of working set pages that have been stolen from real memory. When a stolen page is referenced by the process, a page fault occurs, and the page must be read into memory from paging space.
    Due to the variety of configurations of hardware, software and applications, there is no absolute number to look out for. But five page-ins per second per paging space should be the upper limit. This guideline should not be rigidly adhered to, but used as a reference. This field is important as a key indicator of paging-space activity. If a page-in occurs, there must have been a previous page-out for that page. It is also likely in a memory-constrained environment that each page-in will force a different page to be stolen and, therefore, paged out. But systems could also work fine when they have close to 10 pi per second for 1 minute and then work without any page-ins.
    
  • po
    The po column shows the number (rate) of pages paged out to paging space. Whenever a page of working storage is stolen, it is written to paging space, if it does not yet reside in paging space or if it was modified. If not referenced again, it will remain on the paging device until the process terminates or disclaims the space. Subsequent references to addresses contained within the faulted-out pages results in page faults, and the pages are paged in individually by the system. When a process terminates normally, any paging space allocated to that process is freed. If the system is reading in a significant number of persistent pages (files), you might see an increase in po without corresponding increases in pi. This does not necessarily indicate thrashing, but may warrant investigation into data-access patterns of the applications.
    
  • fr
    Number of pages that were freed per second by the page-replacement algorithm during the interval. As the VMM page-replacement routine scans the Page Frame Table (PFT), it uses criteria to select which pages are to be stolen to replenish the free list of available memory frames. The criteria include both kinds of pages, working (computational) and file (persistent) pages. Just because a page has been freed, it does not mean that any I/O has taken place. For example, if a persistent storage (file) page has not been modified, it will not be written back to the disk. If I/O is not necessary, minimal system resources are required to free a page.
    
  • sr
    Number of pages that were examined per second by the page-replacement algorithm during the interval. The VMM page-replacement code scans the PFT and steals pages until the number of frames on the free list is at least the maxfree value. The page-replacement code might have to scan many entries in the PFT before it can steal enough to satisfy the free list requirements. With stable, unfragmented memory, the scan rate and free rate might be nearly equal. On systems with multiple processes using many different pages, the pages are more volatile and disjoint. In this scenario, the scan rate might greatly exceed the free rate.
    Memory is over committed when the ratio of fr to sr (fr:sr) is high.
    An fr:sr ratio of 1:4 means that for every page freed, four pages had to be examined. It is difficult to determine a memory constraint based on this ratio alone, and what constitutes a high ratio is workload/application dependent.
    
  • cy
    Number of cycles per second of the clock algorithm. The VMM uses a technique known as the clock algorithm to select pages to be replaced. This technique takes advantage of a referenced bit for each page as an indication of what pages have been recently used (referenced). When the page-stealer routine is called, it cycles through the PFT, examining each page's referenced bit.
    The cy column shows how many times per second the page-replacement code has scanned the PFT. Because the free list can be replenished without a complete scan of the PFT and because all of the vmstat fields are reported as integers, this field is usually zero. If not, it indicates a complete scan of the PFT, and the stealer has to scan the PFT again, because fre is still under the maxfree value.
    

One way to determine the appropriate amount of RAM for a system is to look at the largest value for avm as reported by the vmstat command. Multiply that by 4 K to get the number of bytes and then compare that to the number of bytes of RAM on the system. Ideally, avm should be smaller than total RAM. If not, some amount of virtual memory paging will occur. How much paging occurs will depend on the difference between the two values. Remember, the idea of virtual memory is that it gives us the capability of addressing more memory than we have (some of the memory is in RAM and the rest is in paging space). But if there is far more virtual memory than real memory, this could cause excessive paging which then results in delays. If avm is lower than RAM, then paging-space paging could be caused by RAM being filled up with file pages. In that case, tuning the minperm/maxperm values, could reduce the amount of paging-space paging (see
Tuning VMM Page Replacement with the vmtune Command
).
The vmstat -I Command
In operating system versions later than AIX 4.3.3, the vmstat -I (uppercase i) command displays additional information, such as file pages in per-second, file pages out per-second (that is, any VMM page-ins and page-outs that are not paging space page-ins or paging space page-outs). The re and cy columns are not reported with this flag.
The vmstat -s Command
The summary (-s) option sends a summary report to standard output starting from system initialization expressed in absolute counts rather than on an interval basis. The recommended way of using these statistics is to run this command before a workload, save the output, and then run it again after the workload and save its output. The next step is to determine the difference between the two sets of output. An awk script called vmstatit that does this automatically is provided in
Determining Whether the Problem is Related to Disk or Memory
.
# vmstat -s
  3231543 total address trans. faults
    63623 page ins
   383540 page outs
      149 paging space page ins
      832 paging space page outs
        0 total reclaims
   807729 zero filled pages faults
     4450 executable filled pages faults
   429258 pages examined by clock
        8 revolutions of the clock hand
   175846 pages freed by the clock
    18975 backtracks
        0 lock misses
       40 free frame waits
        0 extend XPT waits
    16984 pending I/O waits
   186443 start I/Os
   186443 iodones
141695229 cpu context switches
317690215 device interrupts
        0 software interrupts
        0 traps
55102397 syscalls
The page-in and page-out numbers in the summary represent virtual memory activity to page in or out pages from page space and file space. The paging space ins and outs are representative of only page space. If you are experiencing considerable I/O wait time, but are not memory-constrained, the problem might be due to poor distribution of I/O across drives. To determine if the problem is due to paging space or files, take multiple samples of the vmstat -s command and see if the page ins and outs to paging space are the majority of the total paging. If the system is paging too much, using the vmtune command might help (see
Tuning VMM Page Replacement with the vmtune Command
). Creating separate paging spaces on separate volumes might provide some benefit, but increasing the memory would definitely help.
The ps Command
The ps command can also be used to monitor memory usage of individual processes. The ps v PID command provides the most comprehensive report on memory-related statistics for an individual process, such as:

• Page faults
  
• Size of working segment that has been touched
  
• Size of working segment and code segment in memory
  
• Size of text segment
  
• Size of resident set
  
• Percentage of real memory used by this process

An example:
# ps v
   PID    TTY STAT  TIME PGIN  SIZE   RSS   LIM  TSIZ   TRS %CPU %MEM COMMAND
36626  pts/3 A     0:00    0   316   408 32768    51    60  0.0  0.0 ps v
The most important columns on the resulting ps report are described as follows:
PGIN
Number of page-ins caused by page faults. Since all I/O is classified as page faults, this is basically a measure of I/O volume.
SIZE
Virtual size (in paging space) in kilobytes of the data section of the process (displayed as SZ by other flags). This number is equal to the number of working segment pages of the process that have been touched times 4. If some working segment pages are currently paged out, this number is larger than the amount of real memory being used. SIZE includes pages in the private segment and the shared-library data segment of the process.
RSS
Real-memory (resident set) size in kilobytes of the process. This number is equal to the sum of the number of working segment and code segment pages in memory times 4. Remember that code segment pages are shared among all of the currently running instances of the program. If 26 ksh processes are running, only one copy of any given page of the ksh executable program would be in memory, but the ps command would report that code segment size as part of the RSS of each instance of the ksh program.
TSIZ
Size of text (shared-program) image. This is the size of the text section of the executable file. Pages of the text section of the executable program are only brought into memory when they are touched, that is, branched to or loaded from. This number represents only an upper bound on the amount of text that could be loaded. The TSIZ value does not reflect actual memory usage. This TSIZ value can also be seen by executing the dump -ov command against an executable program (for example, dump -ov /usr/bin/ls).
TRS
Size of the resident set (real memory) of text. This is the number of code segment pages times 4. This number exaggerates memory use for programs of which multiple instances are running. The TRS value can be higher than the TSIZ value because other pages may be included in the code segment such as the XCOFF header and the loader section.
%MEM
Calculated as the sum of the number of working segment and code segment pages in memory times 4 (that is, the RSS value), divided by the size of the real memory of the machine in KB, times 100, rounded to the nearest full percentage point. This value attempts to convey the percentage of real memory being used by the process. Unfortunately, like RSS, it tends the exaggerate the cost of a process that is sharing program text with other processes. Further, the rounding to the nearest percentage point causes all of the processes in the system that have RSS values under 0.005 times real memory size to have a %MEM of 0.0.
Note: The ps command does not indicate memory consumed by shared memory segments or memory-mapped segments. Because many applications use shared memory or memory-mapped segments, the svmon command is a better tool to view the memory usage of these segments.
The svmon Command
The svmon command provides a more in-depth analysis of memory usage. It is more informative, but also more intrusive, than the vmstat and ps commands. The svmon command captures a snapshot of the current state of memory. However, it is not a true snapshot because it runs at the user level with interrupts enabled.
To determine whether svmon is installed and available, run the following command:
# lslpp -lI perfagent.tools
The svmon command can only be executed by the root user.
If an interval is used (-i option), statistics will be displayed until the command is killed or until the number of intervals, which can be specified right after the interval, is reached.
You can use four different reports to analyze the displayed information:
Global (-G)
Displays statistics describing the real memory and paging space in use for the whole system.
Process (-P)
Displays memory usage statistics for active processes.
Segment (-S)
Displays memory usage for a specified number of segments or the top ten highest memory-usage processes in descending order.
Detailed Segment (-D)
Displays detailed information on specified segments.
Additional reports are available in AIX 4.3.3 and later, as follows:
User (-U)
Displays memory usage statistics for the specified login names. If no list of login names is supplied, memory usage statistics display all defined login names.
Command (-C)
Displays memory usage statistics for the processes specified by command name.
Workload Management Class (-W)
Displays memory usage statistics for the specified workload management classes. If no classes are supplied, memory usage statistics display all defined classes.
To support 64-bit applications, the output format of the svmon command was modified in AIX 4.3.3 and later.
Additional reports are available in operating system versions later than 4.3.3, as follows:
Frame (-F)
Displays information about frames. When no frame number is specified, the percentage of used memory is reported. When a frame number is specified, information about that frame is reported.
Tier (-T)
Displays information about tiers, such as the tier number, the superclass name when the -a flag is used, and the total number of pages in real memory from segments belonging to the tier.
How Much Memory is in Use
To print out global statistics, use the -G flag. In this example, we will repeat it five times at two-second intervals.
# svmon -G -i 2 5
       m e m o r y            i n  u s e            p i n        p g  s p a c e
  size inuse  free   pin   work  pers  clnt   work  pers  clnt     size   inuse
16384 16250   134  2006  10675  2939  2636   2006     0     0    40960   12674
16384 16254   130  2006  10679  2939  2636   2006     0     0    40960   12676
16384 16254   130  2006  10679  2939  2636   2006     0     0    40960   12676
16384 16254   130  2006  10679  2939  2636   2006     0     0    40960   12676
16384 16254   130  2006  10679  2939  2636   2006     0     0    40960   12676
The columns on the resulting svmon report are described as follows:
memory
Statistics describing the use of real memory, shown in 4 K pages.
size
Total size of memory in 4 K pages.
inuse
Number of pages in RAM that are in use by a process plus the number of persistent pages that belonged to a terminated process and are still resident in RAM. This value is the total size of memory minus the number of pages on the free list.
free
Number of pages on the free list.
pin
Number of pages pinned in RAM (a pinned page is a page that is always resident in RAM and cannot be paged out).
  
in use
Detailed statistics on the subset of real memory in use, shown in 4 K frames.
work
Number of working pages in RAM.
pers
Number of persistent pages in RAM.
clnt
Number of client pages in RAM (client page is a remote file page).
  
pin
Detailed statistics on the subset of real memory containing pinned pages, shown in 4 K frames.
work
Number of working pages pinned in RAM.
pers
Number of persistent pages pinned in RAM.
clnt
Number of client pages pinned in RAM.
  
pg space
Statistics describing the use of paging space, shown in 4 K pages. This data is reported only if the -r flag is not used. The value reported starting with AIX 4.3.2 is the actual number of paging-space pages used (which indicates that these pages were paged out to the paging space). This differs from the vmstat command in that vmstat's avm column which shows the virtual memory accessed but not necessarily paged out.
size
Total size of paging space in 4 K pages.
inuse
Total number of allocated pages.
  
In our example, there are 16384 pages of total size of memory. Multiply this number by 4096 to see the total real memory size (64 MB). While 16250 pages are in use, there are 134 pages on the free list and 2006 pages are pinned in RAM. Of the total pages in use, there are 10675 working pages in RAM, 2939 persistent pages in RAM, and 2636 client pages in RAM. The sum of these three parts is equal to the inuse column of the memory part. The pin part divides the pinned memory size into working, persistent and client categories. The sum of them is equal to the pin column of the memory part. There are 40960 pages (160 MB) of total paging space, and 12676 pages are in use. The inuse column of memory is usually greater than the inuse column of pg spage because memory for file pages is not freed when a program completes, while paging-space allocation is.
In AIX 4.3.3 and later, systems the output of the same command looks similar to the following:
# svmon -G -i 2 5

               size      inuse       free        pin    virtual
memory        65527      64087       1440       5909      81136
pg space     131072      55824

               work       pers       clnt
pin            5918          0          0
in use        47554      13838       2695


               size      inuse       free        pin    virtual
memory        65527      64091       1436       5909      81137
pg space     131072      55824

               work       pers       clnt
pin            5918          0          0
in use        47558      13838       2695


               size      inuse       free        pin    virtual
memory        65527      64091       1436       5909      81137
pg space     131072      55824

               work       pers       clnt
pin            5918          0          0
in use        47558      13838       2695


               size      inuse       free        pin    virtual
memory        65527      64090       1437       5909      81137
pg space     131072      55824

               work       pers       clnt
pin            5918          0          0
in use        47558      13837       2695


               size      inuse       free        pin    virtual
memory        65527      64168       1359       5912      81206
pg space     131072      55824

               work       pers       clnt
pin            5921          0          0
in use        47636      13837       2695
The additional output field is the virtual field, which shows the number of pages allocated in the system virtual space.
Who is Using Memory?
The following command displays the memory usage statistics for the top ten processes. If you do not specify a number, it will display all the processes currently running in this system.
# svmon -Pau 10

  Pid                         Command        Inuse        Pin      Pgspace
15012                     maker4X.exe         4783       1174         4781
2750                               X         4353       1178         5544
15706                            dtwm         3257       1174         4003
17172                       dtsession         2986       1174         3827
21150                          dtterm         2941       1174         3697
17764                         aixterm         2862       1174         3644
2910                          dtterm         2813       1174         3705
19334                          dtterm         2813       1174         3704
13664                          dtterm         2804       1174         3706
17520                         aixterm         2801       1174         3619

Pid:  15012
Command:  maker4X.exe

Segid  Type  Description          Inuse    Pin  Pgspace  Address Range
1572  pers  /dev/hd3:62              0      0        0  0..-1
  142  pers  /dev/hd3:51              0      0        0  0..-1
1bde  pers  /dev/hd3:50              0      0        0  0..-1
  2c1  pers  /dev/hd3:49              1      0        0  0..7
  9ab  pers  /dev/hd2:53289           1      0        0  0..0
  404  work  kernel extension        27     27        0  0..24580
1d9b  work  lib data                39      0       23  0..607
  909  work  shared library text    864      0        7  0..65535
  5a3  work  sreg[4]                  9      0       12  0..32768
1096  work  sreg[3]                 32      0       32  0..32783
1b9d  work  private               1057      1     1219  0..1306 : 65307..65535
1af8  clnt                         961      0        0  0..1716
    0  work  kernel                1792   1146     3488  0..32767 : 32768..65535
...
The output is divided into summary and detail sections. The summary section lists the top ten highest memory-usage processes in descending order.
Pid 15012 is the process ID that has the highest memory usage. The Command indicates the command name, in this case maker4X.exe. The Inuse column (total number of pages in real memory from segments that are used by the process) shows 4783 pages (each page is 4 KB). The Pin column (total number of pages pinned from segments that are used by the process) shows 1174 pages. The Pgspace column (total number of paging-space pages that are used by the process) shows 4781 pages.
The detailed section displays information about each segment for each process that is shown in the summary section. This includes the segment ID, the type of the segment, description (a textual description of the segment, including the volume name and i-node of the file for persistent segments), number of pages in RAM, number of pinned pages in RAM, number of pages in paging space, and address range.
The Address Range specifies one range for a persistent or client segment and two ranges for a working segment. The range for a persistent or a client segment takes the form '0..x,' where x is the maximum number of virtual pages that have been used. The range field for a working segment can be '0..x : y..65535', where 0..x contains global data and grows upward, and y..65535 contains stack area and grows downward. For the address range, in a working segment, space is allocated starting from both ends and working towards the middle. If the working segment is non-private (kernel or shared library), space is allocated differently. In this example, the segment ID 1b9d is a private working segment; its address range is 0..1306 : 65307..65535. The segment ID 909 is a shared library text working segment; its address range is 0..65535.
A segment can be used by multiple processes. Each page in real memory from such a segment is accounted for in the Inuse field for each process using that segment. Thus, the total for Inuse may exceed the total number of pages in real memory. The same is true for the Pgspace and Pin fields. The sum of Inuse, Pin, and Pgspace of all segments of a process is equal to the numbers in the summary section.
You can use one of the following commands to display the file name associated with the i-node:

• ncheck -i i-node_number volume_name
  
• find file_system_associated_with_lv_name -xdev -inum inode_number -print

To get a similar output in AIX 4.3.3 and later, use the following command:
# svmon -Put 10

------------------------------------------------------------------------------
     Pid Command        Inuse      Pin     Pgsp  Virtual   64-bit    Mthrd
    2164 X              15535     1461    34577    37869        N        N

  Vsid     Esid Type Description           Inuse   Pin Pgsp Virtual Addr Range
  1966        2 work process private        9984     4 31892 32234   0..32272 :
                                                                    65309..65535
  4411        d work shared library text    3165     0 1264  1315   0..65535
     0        0 work kernel seg             2044  1455 1370  4170   0..32767 :
                                                                    65475..65535
  396e        1 pers code,/dev/hd2:18950     200     0    -     -   0..706
  2ca3        - work                          32     0    0    32   0..32783
  43d5        - work                          31     0    6    32   0..32783
  2661        - work                          29     0    0    29   0..32783
  681f        - work                          29     0   25    29   0..32783
  356d        f work shared library data      18     0   18    24   0..310
  34e8        3 work shmat/mmap                2     2    2     4   0..32767
  5c97        - pers /dev/hd4:2                1     0    -     -   0..0
  5575        - pers /dev/hd2:19315            0     0    -     -   0..0
  4972        - pers /dev/hd2:19316            0     0    -     -   0..5
  4170        - pers /dev/hd3:28               0     0    -     -   0..0
  755d        - pers /dev/hd9var:94            0     0    -     -   0..0
  6158        - pers /dev/hd9var:90            0     0    -     -   0..0

------------------------------------------------------------------------------
     Pid Command        Inuse      Pin     Pgsp  Virtual   64-bit    Mthrd
   25336 austin.ibm.    12466     1456     2797    11638        N        N

  Vsid     Esid Type Description           Inuse   Pin Pgsp Virtual Addr Range
  14c3        2 work process private        5644     1  161  5993   0..6550 :
                                                                    65293..65535
  4411        d work shared library text    3165     0 1264  1315   0..65535
     0        0 work kernel seg             2044  1455 1370  4170   0..32767 :
                                                                    65475..65535
  13c5        1 clnt code                    735     0    -     -   0..4424
   d21        - pers /dev/andy:563           603     0    -     -   0..618
   9e6        f work shared library data     190     0    2   128   0..3303
   942        - pers /dev/cache:16            43     0    -     -   0..42
  2ca3        - work                          32     0    0    32   0..32783
  49f0        - clnt                          10     0    -     -   0..471
  1b07        - pers /dev/andy:8568            0     0    -     -   0..0
   623        - pers /dev/hd2:22539            0     0    -     -   0..1
  2de9        - clnt                           0     0    -     -   0..0
  1541        5 mmap mapped to sid 761b        0     0    -     -
  5d15        - pers /dev/andy:487             0     0    -     -   0..3
  4513        - pers /dev/andy:486             0     0    -     -   0..45
   cc4        4 mmap mapped to sid 803         0     0    -     -
  242a        - pers /dev/andy:485             0     0    -     -   0..0
...
The Vsid column is the virtual segment ID, and the Esid column is the effective segment ID. The effective segment ID reflects the segment register that is used to access the corresponding pages.
Detailed Information on a Specific Segment ID
The -D option displays detailed memory-usage statistics for segments.
# svmon -D 404
Segid:  404
Type:  working
Description:  kernel extension
Address Range:  0..24580
Size of page space allocation:  0 pages ( 0.0 Mb)
Inuse:  28 frames ( 0.1 Mb)
    Page   Frame      Pin     Ref     Mod
   12294    3320      pin     ref     mod
   24580    1052      pin     ref     mod
   12293   52774      pin     ref     mod
   24579   20109      pin     ref     mod
   12292   19494      pin     ref     mod
   12291   52108      pin     ref     mod
   24578   50685      pin     ref     mod
   12290   51024      pin     ref     mod
   24577    1598      pin     ref     mod
   12289   35007      pin     ref     mod
   24576     204      pin     ref     mod
   12288     206      pin     ref     mod
    4112   53007      pin             mod
    4111   53006      pin             mod
    4110   53005      pin             mod
    4109   53004      pin             mod
    4108   53003      pin             mod
    4107   53002      pin             mod
    4106   53001      pin             mod
    4105   53000      pin             mod
    4104   52999      pin             mod
    4103   52998      pin             mod
    4102   52997      pin             mod
    4101   52996      pin             mod
    4100   52995      pin             mod
    4099   52994      pin             mod
    4098   52993      pin             mod
    4097   52992      pin     ref     mod
The detail columns are explained as follows:
Page
Specifies the index of the page within the segment.
Frame
Specifies the index of the real memory frame that the page resides in.
Pin
Specifies a flag indicating whether the page is pinned.
Ref
Specifies a flag indicating whether the page's reference bit is on.
Mod
Specifies a flag indicating whether the page is modified.
The size of page space allocation is 0 because all the pages are pinned in real memory.
An example output from AIX 4.3.3 and later, is very similar to the following:
# svmon -D 629 -b

Segid: 629
Type:  working
Address Range: 0..77
Size of page space allocation: 7 pages (  0.0 Mb)
Virtual: 11 frames ( 0.0 Mb)
Inuse: 7 frames ( 0.0 Mb)

           Page      Frame    Pin        Ref        Mod
              0      32304     N          Y          Y
              3      32167     N          Y          Y
              7      32321     N          Y          Y
              8      32320     N          Y          Y
              5      32941     N          Y          Y
              1      48357     N          N          Y
             77      47897     N          N          Y
The -b flag shows the status of the reference and modified bits of all the displayed frames. After it is shown, the reference bit of the frame is reset. When used with the -i flag, it detects which frames are accessed between each interval.
Note: Use this flag with caution because of its performance impacts.
List of Top Memory Usage of Segments
The -S option is used to sort segments by memory usage and to display the memory-usage statistics for the top memory-usage segments. If count is not specified, then a count of 10 is implicit. The following command sorts system and non-system segments by the number of pages in real memory and prints out the top 10 segments of the resulting list.
# svmon -Sau

Segid  Type  Description          Inuse    Pin  Pgspace  Address Range
    0  work  kernel                1990   1408     3722  0..32767 : 32768..65535
    1  work  private, pid=4042     1553      1     1497  0..1907 : 65307..65535
1435  work  private, pid=3006     1391      3     1800  0..4565 : 65309..65535
11f5  work  private, pid=14248    1049      1     1081  0..1104 : 65307..65535
11f3  clnt                         991      0        0  0..1716
  681  clnt                         960      0        0  0..1880
  909  work  shared library text    900      0        8  0..65535
  101  work  vmm data               497    496        1  0..27115 : 43464..65535
  a0a  work  shared library data    247      0      718  0..65535
1bf9  work  private, pid=21094     221      1      320  0..290 : 65277..65535
All output fields are described in the previous examples.
An example output from AIX 4.3.3 and later is similar to the following:
# svmon -Sut 10

  Vsid     Esid Type Description           Inuse   Pin Pgsp Virtual Addr Range
  1966        - work                        9985     4 31892 32234   0..32272 :
                                                                    65309..65535
  14c3        - work                        5644     1  161  5993   0..6550 :
                                                                    65293..65535
  5453        - work                        3437     1 2971  4187   0..4141 :
                                                                    65303..65535
  4411        - work                        3165     0 1264  1315   0..65535
  5a1e        - work                        2986     1   13  2994   0..3036 :
                                                                    65295..65535
  340d        - work misc kernel tables     2643     0  993  2645   0..15038 :
                                                                    63488..65535
  380e        - work kernel pinned heap     2183  1055 1416  2936   0..65535
     0        - work kernel seg             2044  1455 1370  4170   0..32767 :
                                                                    65475..65535
  6afb        - pers /dev/notes:92          1522     0    -     -   0..10295
  2faa        - clnt                        1189     0    -     -   0..2324
Correlating svmon and vmstat Outputs
There are some relationships between the svmon and vmstat outputs. The svmon report of AIX 4.3.2 follows (the example is the same with AIX 4.3.3 and later, although the output format is different):
# svmon -G
       m e m o r y            i n  u s e            p i n        p g  s p a c e
  size inuse  free   pin   work  pers  clnt   work  pers  clnt     size   inuse
16384 16254   130  2016  11198  2537  2519   2016     0     0    40960   13392
The vmstat command was run in a separate window while the svmon command was running. The vmstat report follows:
# vmstat 5
kthr     memory             page              faults        cpu
----- ----------- ------------------------ ------------ -----------
r  b   avm   fre  re  pi  po  fr   sr  cy  in   sy  cs us sy id wa
0  0 13392   130   0   0   0   0    2   0 125  140  36  2  1 97  0
0  0 13336   199   0   0   0   0    0   0 145 14028  38 11 22 67  0
0  0 13336   199   0   0   0   0    0   0 141   49  31  1  1 98  0
0  0 13336   199   0   0   0   0    0   0 142   49  32  1  1 98  0
0  0 13336   199   0   0   0   0    0   0 145   49  32  1  1 99  0
0  0 13336   199   0   0   0   0    0   0 163   49  33  1  1 92  6
0  0 13336   199   0   0   0   0    0   0 142   49  32  0  1 98  0
The global svmon report shows related numbers. The vmstatfre column relates to the svmon memory free column. The number that vmstat reports as Active Virtual Memory (avm) is reported by the svmon command as pg space inuse (13392).
The vmstat avm column provides the same figures as the pg space inuse column of the svmon command except starting with AIX 4.3.2 where Deferred Page Space Allocation is used. In that case, the svmon command shows the number of pages actually paged out to paging space whereas the vmstat command shows the number of virtual pages accessed but not necessarily paged out (see
Looking at Paging Space and Virtual Memory
).
Correlating svmon and ps Outputs
There are some relationships between the svmon and ps outputs. The svmon report of AIX 4.3.2 follows (the example is the same with AIX 4.3.3 and later, although the output format is different):
# svmon -P 7226

  Pid                         Command        Inuse        Pin      Pgspace
7226                         telnetd          936          1           69

Pid:   7226
Command:  telnetd

Segid  Type  Description          Inuse    Pin  Pgspace  Address Range
  828  pers  /dev/hd2:15333           0      0        0  0..0
1d3e  work  lib data                 0      0       28  0..559
  909  work  shared library text    930      0        8  0..65535
1cbb  work  sreg[3]                  0      0        1  0..0
1694  work  private                  6      1       32  0..24 : 65310..65535
12f6  pers  code,/dev/hd2:69914      0      0        0  0..11
Compare with the ps report, which follows:
# ps v 7226
   PID    TTY STAT  TIME PGIN  SIZE   RSS   LIM  TSIZ   TRS %CPU %MEM COMMAND
  7226      - A     0:00   51   240    24 32768    33     0  0.0  0.0 telnetd
SIZE refers to the virtual size in KB of the data section of the process (in paging space). This number is equal to the number of working segment pages of the process that have been touched (that is, the number of paging-space pages that have been allocated) times 4. It must be multiplied by 4 because pages are in 4 K units and SIZE is in 1 K units. If some working segment pages are currently paged out, this number is larger than the amount of real memory being used. The SIZE value (240) correlates with the Pgspace number from the svmon command for private (32) plus lib data (28) in 1 K units.
RSS refers to the real memory (resident set) size in KB of the process. This number is equal to the sum of the number of working segment and code segment pages in memory times 4. Remember that code segment pages are shared among all of the currently running instances of the program. If 26 ksh processes are running, only one copy of any given page of the ksh executable program would be in memory, but the ps command would report that code segment size as part of the RSS of each instance of the ksh program. The RSS value (24) correlates with the Inuse numbers from the svmon command for private (6) working-storage segments, for code (0) segments, and for lib data (0) of the process in 1-K units.
TRS refers to the size of the resident set (real memory) of text. This is the number of code segment pages times four. As was noted earlier, this number exaggerates memory use for programs of which multiple instances are running. This does not include the shared text of the process. The TRS value (0) correlates with the number of the svmon pages in the code segment (0) of the Inuse column in 1 K units. The TRS value can be higher than the TSIZ value because other pages, such as the XCOFF header and the loader section, may be included in the code segment.
The following calculations can be made for the values mentioned:
SIZE = 4 * Pgspace of (work lib data + work private)
RSS  = 4 * Inuse of (work lib data + work private + pers code)
TRS  = 4 * Inuse of (pers code)
Calculating the Minimum Memory Requirement of a Program
To calculate the minimum memory requirement of a program, the formula would be:
Total memory pages (4 KB units) = T + ( N * ( PD + LD ) ) + F
where:
T
= Number of pages for text (shared by all users)
N
= Number of copies of this program running simultaneously
PD
= Number of working segment pages in process private segment
LD
= Number of shared library data pages used by the process
F
= Number of file pages (shared by all users)
Multiply the result by 4 to obtain the number of kilobytes required. You may want to add in the kernel, kernel extension, and shared library text segment values to this as well even though they are shared by all processes on the system. For example, some applications like CATIA and databases use very large shared library modules. Note that because we have only used statistics from a single snapshot of the process, there is no guarantee that the value we get from the formula will be the correct value for the minimum working set size of a process. To get working set size, one would need to run a tool such as the rmss command or take many snapshots during the life of the process and determine the average values from these snapshots (see
Assessing Memory Requirements Through the rmss Command
).
If we estimate the minimum memory requirement for the program pacman, shown in
Finding Memory-Leaking Programs
, the formula would be:
T
= 2 (Inuse of code,/dev/lv01:12302 of pers)
PD
= 1632 (Inuse of private of work)
LD
= 12 (Inuse of lib data of work)
F
= 1 (Inuse of /dev/hd2:53289 of pers
That is: 2 + (N * (1632+ 12)) + 1, equal to 1644 * N + 3 in 4 KB units.


本文来自ChinaUnix博客，如果查看原文请点：http://blog.chinaunix.net/u/2984/showart_102785.html