nicht angemeldet
perl benchmark
Hallo,
ich habe da ein kleines Verständnisproblem. Bei einem
Perl Skript (über Apache und Netscape) mit dem "timediff"
Aufruf erhalte ich folgendes Ergebnis:
The code took: 5 wallclock secs ( 2.37 usr + 0.00 sys = 2.37 CPU)
Sollte "wallclock" nicht in etwa die Summe von "usr" und "sys" sein ?
Was bedeutet nun "wallclock" und wo liegt der Unterschied
zwischen "usr" und "sys" ?
Leider habe ich hierzu nirgends eine Erklärung gefunden. Ich weiß ;-)
sicherlich gibt es die, aber wo ??? Perldoc und google haben jedenfalls
nicht geholfen ?
Vielen Dank
Gruß
Jean-Luc
-
hi!
The code took: 5 wallclock secs ( 2.37 usr + 0.00 sys = 2.37 CPU)
Sollte "wallclock" nicht in etwa die Summe von "usr" und "sys" sein?Nicht unbedingt. Ich vermute eher, dass "wallclock" die gesamte reale
Laufzeit des Programms ist. Wenn also nebenbei noch andere Prozesse
laufen, erhalten die natürlich auch Laufzeit, die dann nicht bei der
sys- und usr-Zeit des Prozesses mitgerechnet wird.Was bedeutet nun "wallclock" und wo liegt der Unterschied zwischen
"usr" und "sys" ?Hm, "usr" ist die CPU-Zeit, die für die Ausführung der Instruktionen
benötigt wird. "sys" ist dann vermutlich die Zeit, die der Prozess
im supervisor-Mode verbringt, wenn er zb. Syscalls macht oder wenn
er initialisiert wird. Dabei kann ich mich aber auch irren... ;)bye, Frank!
-
Hallo,
Hm, "usr" ist die CPU-Zeit, die für die Ausführung der
Instruktionen benötigt wird. "sys" ist dann vermutlich die Zeit,
die der Prozess im supervisor-Mode verbringt, wenn er zb.
Syscalls macht oder wenn er initialisiert wird. Dabei kann ich
mich aber auch irren... ;)Ich denke eher, 'usr' ist die User-Zeit und 'sys' ist die System-
Zeit. User-Zeit heisst hierbei die reale Zeit, die der User warten
muss. Daraus ergeben sich die 'Wallclocks', die wirklich reale Zeit,
die der Prozess gelaufen ist. Dafuer wuerde der folgende Abschnitt
aus `perldoc Benchmark' und die allgemeine Verwendung dieser
Begriffe sprechen:NOTES
The data is stored as a list of values from the time and
times functions:($real, $user, $system, $children_user, $children_system)
in seconds for the whole loop (not divided by the number
of rounds).Gruesse,
CK-
Hallo,
Hm, "usr" ist die CPU-Zeit, die für die Ausführung der Instruktionen benötigt wird. "sys" ist dann vermutlich die Zeit, die der Prozess im supervisor-Mode verbringt, wenn er zb. Syscalls macht oder wenn er initialisiert wird. Dabei kann ich mich aber auch irren... ;)
Ich denke eher, 'usr' ist die User-Zeit und 'sys' ist die System- Zeit. User-Zeit heisst hierbei die reale Zeit, die der User warten muss. Daraus ergeben sich die 'Wallclocks', die wirklich reale Zeit, die der Prozess gelaufen ist. Dafuer wuerde der folgende Abschnitt aus `perldoc Benchmark' und die allgemeine Verwendung dieser Begriffe sprechen:
NOTES The data is stored as a list of values from the time and times functions:
($real, $user, $system, $children_user, $children_system)
in seconds for the whole loop (not divided by the number of rounds).
Gruesse, CK
Neeneeneee!
USR ist die Zeit, die ein Prozess im User-Mode, also außerhalb des Kernels, läuft. SYS ist die Zeit, die der Prozess im System-Mode läuft. Und Wallclock ist die Zeit zwischen Start und Ende des Prozesses, so wie ein Außenstehender (der Benutzer) die Zeit (mit der Uhr an der Wand) gemessen hätte.
Das hat übrigens nix mit Perl zu tun, das sind einfach API-Funktionen der C-Library, die auf die Prozessdaten zugreifen.
Wen's interessiert, hier die Man-Pages aus Linux.
Alexander
TIME(1) TIME(1)
NAME time - time a simple command or give resource usage
SYNOPSIS time [options] command [arguments...]
DESCRIPTION The time command runs the specified program command with the given arguments. When command finishes, time writes a message to standard output giving timing statistics about this program run. These statistics consist of (i) the elapsed real time between invocation and termination, (ii) the user CPU time (the sum of the tms_utime and tms_cutime values in a struct tms as returned by times(2)), and (iii) the system CPU time (the sum of the tms_stime and tms_cstime values in a struct tms as returned by times(2)).
OPTION -p When in the POSIX locale, use the precise tradi- tional format "real %f\nuser %f\nsys %f\n" (with numbers in seconds) where the number of deci- mals in the output for %f is unspecified but is sufficient to express the clock tick accuracy, and at least one.
ENVIRONMENT The variables LANG, LC_ALL, LC_CTYPE, LC_MESSAGES, LC_NUMERIC, NLSPATH and PATH are used. The last one to search for command. The remaining ones for the text and formatting of the output.
EXIT STATUS If command was invoked, the exit status is that of com- mand. Otherwise it is 127 if command could not be found, 126 if it could be found but could not be invoked, and some other nonzero value (1-125) if something else went wrong.
SEE ALSO times(2),
GNU VERSION Below a description of the GNU 1.7 version of time. Dis- regarding the name of the utility, GNU makes it output lots of useful information, not only about time used, but also on other resources like memory, I/O and IPC calls (where available). The output is formatted using a format string that can be specified using the -f option or the TIME environment variable.
The default format string is %Uuser %Ssystem %Eelapsed %PCPU (%Xtext+%Ddata %Mmax)k %Iinputs+%Ooutputs (%Fmajor+%Rminor)pagefaults %Wswaps
When the -p option is given the (portable) output format real %e user %U sys %S is used.
The format string The format is interpreted in the usual printf-like way. Ordinary characters are directly copied, tab, newline and backslash are escaped using \t, \n and \, a percent sign is represented by %%, and otherwise % indicates a conver- sion. The program time will always add a trailing newline itself. The conversions follow. All of those used by tcsh(1) are supported.
Time
%E Elapsed real time (in [hours:]minutes:seconds).
%e (Not in tcsh.) Elapsed real time (in seconds).
%S Total number of CPU-seconds that the process spent in kernel mode.
%U Total number of CPU-seconds that the process spent in user mode.
%P Percentage of the CPU that this job got, computed as (%U + %S) / %E.
Memory
%M Maximum resident set size of the process during its lifetime, in Kbytes.
%t (Not in tcsh.) Average resident set size of the process, in Kbytes.
%K Average total (data+stack+text) memory use of the process, in Kbytes.
%D Average size of the process's unshared data area, in Kbytes.
%p (Not in tcsh.) Average size of the process's unshared stack space, in Kbytes.
%X Average size of the process's shared text space, in Kbytes.
%Z (Not in tcsh.) System's page size, in bytes. This is a per-system constant, but varies between sys- tems.
%F Number of major page faults that occurred while the process was running. These are faults where the page has to be read in from disk.
%R Number of minor, or recoverable, page faults. These are faults for pages that are not valid but which have not yet been claimed by other virtual pages. Thus the data in the page is still valid but the system tables must be updated.
%W Number of times the process was swapped out of main memory.
%c Number of times the process was context-switched involuntarily (because the time slice expired).
%w Number of waits: times that the program was con- text-switched voluntarily, for instance while wait- ing for an I/O operation to complete.
I/O
%I Number of file system inputs by the process.
%O Number of file system outputs by the process.
%r Number of socket messages received by the process.
%s Number of socket messages sent by the process.
%k Number of signals delivered to the process.
%C (Not in tcsh.) Name and command line arguments of the command being timed.
%x (Not in tcsh.) Exit status of the command.
GNU OPTIONS -f FORMAT, --format=FORMAT Specify output format, possibly overriding the for- mat specified in the environment variable TIME.
-p, --portability Use the portable output format.
-o FILE, --output=FILE Do not send the results to stderr, but overwrite the specified file.
-a, --append (Used together with -o.) Do not overwrite but append.
-v, --verbose Give very verbose output about all the program knows about.
GNU STANDARD OPTIONS --help Print a usage message on standard output and exit successfully.
-V, --version Print version information on standard output, then exit successfully.
-- Terminate option list.
BUGS Not all resources are measured by all versions of Unix, so some of the values might be reported as zero. The present selection was mostly inspired by the data provided by 4.2 or 4.3BSD.
GNU time version 1.7 is not yet localized. Thus, it does not implement the POSIX requirements.
The environment variable TIME was badly chosen. It is not unusual for systems like autoconf or make to use environ- ment variables with the name of a utility to override the utility to be used. Uses like MORE or TIME for options to programs (instead of program path names) tend to lead to difficulties.
It seems unfortunate that -o overwrites instead of appends. (That is, the -a option should be the default.)
Mail suggestions and bug reports for GNU time to bug-utils@prep.ai.mit.edu Please include the version of time , which you can get by running time --version and the operating system and C compiler you used.
SEE ALSO tcsh(1), times(2), wait3(2)
AUTHORS David Keppel Original version
David MacKenzie POSIXization, autoconfiscation, GNU getopti- zation, documentation, other bug fixes and improvements.
Arne Henrik Juul Helped with portability
Francois Pinard Helped with portability
2000-12-11 TIME(1) TIMES(2) Linux Programmer's Manual TIMES(2)
NAME times - get process times
SYNOPSIS #include <sys/times.h>
clock_t times(struct tms *buf);
DESCRIPTION The times() function stores the current process times in the struct tms that buf points to. The struct tms is as defined in <sys/times.h>:
struct tms { clock_t tms_utime; /* user time / clock_t tms_stime; / system time / clock_t tms_cutime; / user time of children / clock_t tms_cstime; / system time of children */ };
The tms_utime field contains the CPU time spent executing instructions of the calling process. The tms_stime field contains the CPU time spent in the system while executing tasks on behalf of the calling process. The tms_cutime field contains the sum of the tms_utime and tms_cutime values for all waited-for terminated children. The tms_cstime field contains the sum of the tms_stime and tms_cstime values for all waited-for terminated children.
Times for terminated children (and their descendants) is added in at the moment wait(2) or waitpid(2) returns their process ID. In particular, times of grandchildren that the children did not wait for are never seen.
All times reported are in clock ticks.
RETURN VALUE The function times returns the number of clock ticks that have elapsed since an arbitrary point in the past. For Linux this point is the moment the system was booted. This return value may overflow the possible range of type clock_t. On error, (clock_t) -1 is returned, and errno is set appropriately.
NOTE The number of clock ticks per second can be obtained using sysconf(_SC_CLK_TCK); In POSIX-1996 the symbol CLK_TCK (defined in <time.h>) is mentioned as obsolescent. It is obsolete now.
Note that clock(3) returns values of type clock_t that are not measured in clock ticks but in CLOCKS_PER_SEC.
CONFORMING TO SVr4, SVID, POSIX, X/OPEN, BSD 4.3
HISTORICAL NOTES SVr1-3 returns long and the struct members are of type time_t although they store clock ticks, not seconds since the epoch. V7 used long for the struct members, because it had no type time_t yet.
On older systems the number of clock ticks per second is given by the variable HZ.
SEE ALSO time(1), getrusage(2), wait(2), clock(3), sysconf(3)
Linux 2000-12-11 TIMES(2)
-
-