XFS, ReiserFS, And ext3 Comparisons

[Kurt Wall]

An unnamed Administration source, Condon Thomas A KPWA, wrote:
% 
% Net Llama! wrote:
% > Last week there was a thread on the Linux kernel mailng list comparing
% > XFS, reiserFS & ext3:
% > http://kt.zork.net/kernel-traffic/latest.html#13
% > 
% > looks like ext3 came in last, resierFS first, XFS in the middle.
% 
% Could someone help me interpret these numbers?
% 
% ********************************************
% StPeter:/mnt/part1# time cp -rf /usr/src/kernel-source-2.4.20 ./
% 
% real 1m3.501s
% user 0m0.140s
% sys 0m2.680s
% 
% StPeter:/mnt/part2# time cp -rf /usr/src/kernel-source-2.4.20 ./
% 
% real 0m3.696s *************** so fast...
% user 0m0.110s
% sys 0m3.570s
% 
% StPeter:/mnt/part3# time cp -rf /usr/src/kernel-source-2.4.20 ./
% 
% real 1m29.697s
% user 0m0.090s
% sys 0m2.490s
% ********************************************
% 
% It is my understanding that "real" is the total time it took to process the
% command, "user" is the time it took to process the user's portion of the
% command, and "sys" is the time it took to process the system's portion of
% the command.  I *thought* that "real" was affected by what *else* the cpu
% might be doing at the time, so that the true measure was "user" + "sys" =
% processing time used.  Am I wrong in this?

"real" is the elapsed time, a/k/a "wall clock time"
"user" is the time spent executing user mode code
"sys" is the time spent executing system calls

What you don't see is the impact of caches, time spent in block
transfers, that sort of thing.

Kurt
-- 
A journey of a thousand miles begins with a cash advance.