.\" $Id$
.\"
.TH ubench 8 "21 February 2000" "ubench 0.3" "Linux System Manager's Manual"

.SH NAME
ubench \- Unix Benchmark Utility

.SH SYNOPSIS
.in +.5i
.ti -.5i
ubench
\%[\-cmhs]
.in -.5i
.SH DESCRIPTION
.B ubench
is an attempt to introduce a single measure of
perfomance among machines running various flavors of Unix.

.B Ubench
is executing rather senseless mathematical
integer and floating-point calculations for 3 mins
concurrently using several processes, and the result
is
.B Ubench CPU
benchmark. The ratio of floating-point calculations
to integer is about 1:3.
.B Ubench
will spawn about 2 concurrent processes for each CPU
available on the system. This ensures all available
raw CPU horsepower is used.

.B Ubench
is executing rather senseless memory allocation and
memory to memory copying operations for another 3 mins
concurrently using several processes, and the result
is
.B Ubench MEM
benchmark.

.B Ubench
will take an average of
.B CPU
and
.B MEM
benchmarks and calculate
.B Ubench AVG
benchmark which is supposed to represent the true computing
power of the system.

.SH OPTIONS
.TP
.BI \-c
With this flag
.B ubench
will only execute
.B CPU
benchmark.
.TP
.BI \-m
With this flag
.B ubench
will only execute
.B MEM
benchmark.
.TP
.BI \-s
Prevents
.B ubench
from spawning multiple processes. This
will force ubench to use only 1 CPU on the 
system and the result will be
.B Ubench Single CPU
and
.B Ubench Single MEM
benchmarks.
.TP
.BI \-h
Prints usage information.

.SH FILES
.PD 0
The documentation on
.B ubench
and current list of
.B ubench
benchmarks are available at
.LP
.B http://www.phystech.com/download/ubench.html

.SH BUGS
.PD 0
The ratio which
.B CPU
and
.B MEM
scores contribute to the
.B CPU
and
.B MEM
benchmarks,
and to the final
.B AVG
benchmark has been calculated so that
.B CPU
and
.B MEM
scores averaged over several widely different
computer systems are equal to each other
and make equal contribution to average
.B AVG
score. As more and more benchmarks are available, the ratio
will be calculated more precisely.

.SH AUTHORS
.LP
Sergei Viznyuk <sv@phystech.com>