root / synthbench / bonnie++ / zcav.8 @ 0:839f52ef7657
History | View | Annotate | Download (2.9 kB)
1 |
.TH zcav 8 |
---|---|
2 |
.SH "NAME" |
3 |
zcav \- program to test raw hard drive throughput. |
4 |
|
5 |
.SH "SYNOPSIS" |
6 |
.B zcav |
7 |
.I [\-b block\-size[:chunk\-size]] [\-c count] [\-n num_blocks] [\-w] |
8 |
.I [\-u uid\-to\-use:gid\-to\-use] [\-g gid\-to\-use] |
9 |
.I [\-f] file\-name |
10 |
|
11 |
.SH "DESCRIPTION" |
12 |
This manual page documents briefly the |
13 |
.BR zcav , |
14 |
program. |
15 |
.P |
16 |
Modern hard drives have a constant rotational speed but have varying numbers |
17 |
of sectors per track (outside tracks are longer and have more sectors). This |
18 |
is referred to as Zoned Constant Angular Velocity (or ZCAV). The outer tracks |
19 |
will have a higher data transfer rate due to having more sectors per track, |
20 |
these tracks generally have the lower track/sector numbers. |
21 |
.P |
22 |
This program tests the ZCAV performance of a hard drive, by reading the entire |
23 |
data on it a specified number of times. The file name given as the first |
24 |
parameter, it can be specified as |
25 |
.BR \- , |
26 |
for standard input. This file will be opened as read\-only and in usual |
27 |
operation it will be |
28 |
.BR /dev/hdX |
29 |
or |
30 |
.BR /dev/ide/host0/busX/targetY/lun0/disc |
31 |
depending on whether you use devfs or not (NB operating systems other than |
32 |
Linux will have different device names). |
33 |
.P |
34 |
The output should be able to be easily graphed with |
35 |
.BR gnuplot |
36 |
which is what I use to view the results. |
37 |
|
38 |
.SH "OPTIONS" |
39 |
|
40 |
.TP |
41 |
.B \-b |
42 |
the size (in Meg) of the blocks to read/write (default 100M), optionally |
43 |
followed by a ':' and the chunk size for read/write operations (default 1M). |
44 |
Note that the chunk size must be less than or equal to the block size and |
45 |
must also be significantly less than the size of the RAM in the machine. |
46 |
Also note that for the write test there will be a fsync() after writing every |
47 |
chunk. |
48 |
|
49 |
.TP |
50 |
.B \-c |
51 |
the number of times to read/write the entire disk. |
52 |
|
53 |
.TP |
54 |
.B \-n |
55 |
the amount of data (in Meg) to read/write on each pass (default the entire device). |
56 |
|
57 |
.TP |
58 |
.B \-f |
59 |
the file\-name for the input data. This isn't needed on well configured |
60 |
systems that have a recent Glibc where you can specify the file name without |
61 |
the \-f flag. |
62 |
|
63 |
.TP |
64 |
.B \-u |
65 |
user\-id to use. When running as root specify the UID to run the tests as, it |
66 |
is not recommended to use root, so if you want to run as root use |
67 |
.B \-u root. |
68 |
Also if you want to specify the group to run as then use the |
69 |
.B user:group |
70 |
format. If you specify a user by name but no group then the primary group of |
71 |
that user will be chosen. If you specify a user by number and no group then |
72 |
the group will be |
73 |
.B nogroup. |
74 |
|
75 |
.TP |
76 |
.B \-g |
77 |
group\-id to use. Same as using |
78 |
.B :group |
79 |
for the |
80 |
.B \-u |
81 |
parameter, just a different way to specify it for compatibility with other |
82 |
programs. |
83 |
|
84 |
.TP |
85 |
.B \-w |
86 |
write zero blocks to the disk instead of reading from the disk - will destroy data! |
87 |
|
88 |
.SH "AUTHOR" |
89 |
This program, it's manual page, and the Debian package were written by |
90 |
Russell Coker <russell@coker.com.au>. |
91 |
|
92 |
.SH "AVAILABILITY" |
93 |
The source is available from http://www.coker.com.au/bonnie++ . |
94 |
.P |
95 |
See http://etbe.coker.com.au/category/benchmark for further information. |