io transfer test

io transfer test

Post by hopehope_1 » Wed, 15 Sep 2004 21:01:59

Hi ,

I try to understand the max-io-size of my linux systems .

Here is my config:

Server1: 4 -Itanium64 cpus , 8gb.ram , ext2 and ocfs file systems ,
emc , raid10 , redhat advanced 2.1

Server2 : 4 intel xeon cpus , 4gb.ram , ext3 file system.

This is my small test program:

#include <fcntl.h>
FILE *f ;
int i,size;
struct statfs st;
char buffer[4194304];
printf("filetype:%X,optimal transfer block size:%d,number of
in file system:%d,free blocks in os :%
i = open("t.ykb",O_RDWR,0600);
printf("bytes read:%d\n",size);

I compile this code by using :

gcc t.c -o t

and run strace with time output:

strace -T t

First of all , can this program measure the max-io-size on linux? Is
the logic correct?

Thinkng that the first answer is ok , here is the problem:

On itanium box , the code fails by dumping core if i try to read
something > 7mb.

On xeon box it fails if i read something > 16mb.

This is the strace output of xeon box:

statfs("t.ykb", {f_type="EXT2_SUPER_MAGIC", f_bsize=4096,
f_blocks=83436880, f_bfree=75315827, f_bavail=71077479,
f_files=42385408, f_ffree=42299381, f_fsid={0, 0}, f_namelen=255,
f_frsize=0}) = 0 <0.000431>
fstat64(1, {st_mode=S_IFCHR|0620, st_rdev=makedev(136, 0), ...}) = 0
0) = 0xb75e6000 <0.000015>
write(1, "filetype:EF53,optimal transfer b"...,
116filetype:EF53,optimal transfer block size:4096,number of blocks in
file system:83436880,free blocks in os :75315827
) = 116 <0.000018>
open("t.ykb", O_RDWR) = 3 <0.000015>
write(1, "errno:0\n", 8errno:0
) = 8 <0.000016>
read(3, "CEKILIS_GRUP_KOD|VERDE_KOD|CEKIL"..., 9000000) = 9000000
write(1, "bytes read:9000000\n", 19bytes read:9000000
) = 19 <0.000019>

Does this mean that the os read 9MB. at once? Do you think such a big
value is possible?

I need to measure this value in order to reconfigure my oracle
database disk architecture.

Kind Regards,