06-28-2014
How to measure disk IO 5.0.7? (sar, return values are not valid)
Hello
I am analyzing disk performance OSR5.0.7 running inside VirtualBox.
GUEST: osr5.0.7; 1GB ram; raw disk
HOST: SLES11SP3, 4GB ram; 1 disc SATA2-7200rpm
But I'm not sure how to do it right (the values returned by sar not match the values of the physical machine)
The attributes analyzed are:
blks/s: if not int or out, but converted to MB/s should be: blks/s * 512 /1024 /1024
bread/s: I understand you are the KB/s read from the disk that are not in the cache, with this iconclude that it is the ratio of the hard disk read.
bwrit/s: I understand you are the KB/s written when a flush is done, so i conclude that it is the ratio of the hard disk write.
To measure the performance I use the command:
time sh -c "dd if=/dev/zero of=pruebaHD-4254.dd count=131072 bs=8k ibs=8k conv=sync && sync"
Generating a file the size of the ram and then calculate the MB/s based on the value returned by time (repeat the process with different sizes of bs, then iterate 3 times)
With that understanding, when I see I have the values:
blks/s : max: 86.98 MB/s avg: 16.77 MB/s
bread/s : max: 3.10 MB/s avg: 241.66 KB/s
bwrit/s : max: 86.98 MB/s avg: 16.12 MB/s
According to "time dd" achievement Min: 20MB/s Max: 43MB/s Avg: 30MB/s.
According zabbix host 36MB/s
Since the values are not similar, they conclude that something is wrong in my analysis.
I am misunderstanding sar values?
It is a misconception and I'm looking at it all wrong?
How I can know the ratio I / O disk presicion in OSR5.0.7?
thanks
10 More Discussions You Might Find Interesting
1. UNIX for Dummies Questions & Answers
Sys: HP-UX 9000
In the calling script how do I 'read' the return/exit value of a called script?:confused:
THX in advance for any assistence.:) (1 Reply)
Discussion started by: vslewis
1 Replies
2. Shell Programming and Scripting
I think the $? returns 0 if the last issued command was successful and otherwise if not. But does anyone knows the value list that may be returned ? (or it is only zero/one ? )
Thanks in advance,
Abrahao. (3 Replies)
Discussion started by: 435 Gavea
3 Replies
3. Shell Programming and Scripting
I'm using the following in my script.
if echo $cpuidle |/usr/bin/egrep ; then
when I issue this statement it issues the value of the variable back to stdout which ends up in my output file.
Is there a better way to write this?
I'm using ksh on solaris 9. (3 Replies)
Discussion started by: MizzGail
3 Replies
4. UNIX for Dummies Questions & Answers
Hi,
I got a request from a client to check on their server's cpu, memory and disk utilization. Am planning to use the sar and iostat commands to achieve this. The server is used for sending and receiving messages between the stockbrokers, so I was thinking to run the commands during the market... (9 Replies)
Discussion started by: 60doses
9 Replies
5. Shell Programming and Scripting
Hi.
I have a script like this:
nawk 'BEGIN {FS=","; TOT1=0; REJ1=0;} {
if($7=="TOTAL") { TOT1=TOT1 +$8}
if($7=="REJS") { REJ1=REJ1 +$8}
}' FILE_123.dat
and... (1 Reply)
Discussion started by: mrodrig
1 Replies
6. Shell Programming and Scripting
Hi,
I have a script which does something like the below:
execute_some_script.sh $arg1 $arg2 `exec-some-cmd`
if then;
do something
else
do something else
fi
However, during some cases, there is an error saying:
line xxx: [: too many arguments
at the line number which has... (5 Replies)
Discussion started by: laloo
5 Replies
7. Shell Programming and Scripting
Hi,
I need a shell script, which would search the result values from another files.
1)execute " select column1 from table_name" query on the table.
2)Based on the result, need to be grep from .wft files.
could please explain about this.Below is the way i am using.
#!/bin/sh... (4 Replies)
Discussion started by: Rami Reddy
4 Replies
8. Shell Programming and Scripting
Gents,
Here again,, I have a problem to get a extract of some values.
Every 2 rows I got a new number ( columns 12-25 ), I mean the values are repeated every 2 rows can be 3 times or 4 sometimes,, to identify that there is duplicate values a index value are write in colunm 26 ( increase 1,2,... (4 Replies)
Discussion started by: jiam912
4 Replies
9. UNIX for Advanced & Expert Users
Hello,
I have rebooted the RHEL VM but after rebooting the vm it not showing all the partition mounted on OS level, if i'll execute the fdisk -l command, then i'm able to see the same disk. below is the fdisk output :
# fdisk -l
Disk /dev/sda: 107.6 GB, 107639996416 bytes
255 heads, 63... (1 Reply)
Discussion started by: purushottamaher
1 Replies
10. Shell Programming and Scripting
Hi,
Please have a look at the look at the below top and sar commands.
top -bn1 | grep load | awk '{printf "CPU Load: %.2f\n", $(NF-2)}'
CPU Load: 0.52
sar -u 1 1
Linux 3.10.0-514.16.1.el7.x86_64 (mymac) 06/01/2017 _x86_64_ (2 CPU)
03:27:40 PM CPU %user %nice ... (1 Reply)
Discussion started by: mohtashims
1 Replies
RL(4) Kernel Interfaces Manual RL(4)
NAME
rl - RL-11/RL01, RL02 moving-head disk
SYNOPSIS
/sys/conf/SYSTEM:
NRL rl_drives # RL01/02
/etc/dtab:
#Name Unit# Addr Vector Br Handler(s) # Comments
rl ? 174400 160 5 rlintr # rl-01/02
major device number(s):
raw: 16
block: 7
minor device encoding:
bits 0007 specify partition of RL drive
bits 0070 specify RL drive
DESCRIPTION
The block files access the disk via the system's normal buffering mechanism and may be read and written without regard to physical disk
records. There is also a `raw' interface which provides for direct transmission between the disk and the user's read or write buffer. A
single read or write call results in exactly one I/O operation and therefore raw I/O is considerably more efficient when many words are
transmitted. The names of the raw files conventionally begin with an extra `r.'
In raw I/O the buffer must begin on a word (even) boundary, and counts should be a multiple of 512 bytes (a disk sector). Likewise seek
calls should specify a multiple of 512 bytes.
DISK SUPPORT
The RL01 drives are each 10240 blocks long and the RL02 drives are 20480 blocks long.
On a RL02 there is room for a full sized root ('a') partition and a reasonable sized swap ('b') partition. The RL01 can only (realisti-
cally) have a single 5mb partition.
FILES
/dev/rl[0-3][a-h] block files
/dev/rrl[0-3][a-h] raw files
/dev/MAKEDEV script to create special files
/dev/MAKEDEV.local script to localize special files
SEE ALSO
hk(4), ra(4), ram(4), rk(4), rp(4), rx(4), si(4), xp(4), dtab(5), autoconfig(8)
DIAGNOSTICS
rl%d: hard error sn%d cs=%b da=%b. An unrecoverable error occurred during transfer of the specified sector of the specified disk. The
contents of the two error registers are also printed in octal and symbolically with bits decoded. The error was either unrecoverable, or a
large number of retry attempts could not recover the error.
rl%d: hard error sn%d mp=%b da=%b. An unrecoverable drive error occured during transfer of the specified sector of the specified disk.
The contents of the two error registers are also printed in octal and symbolically with bits decoded. The error was either unrecoverable,
or a large number of retry attempts could not recover the error.
rl%d: write locked. The write protect switch was set on the drive when a write was attempted. The write operation is not recoverable.
rl%d: can't get status. A ``get status'' command on the specified drive failed. The error is unrecoverable.
BUGS
In raw I/O read and write(2) truncate file offsets to 512-byte block boundaries, and write scribbles on the tail of incomplete blocks.
Thus, in programs that are likely to access raw devices, read, write and lseek(2) should always deal in 512-byte multiples.
DEC-standard error logging should be supported.
A program to analyze the logged error information (even in its present reduced form) is needed.
3rd Berkeley Distribution August 20, 1987 RL(4)