Here are some pretty surface area graphs from the EC2 benchmark, the stepping down indicates from CPU cache to Memory cache, the last cliff is down to disk once the file was larger the available memory.
As I mentioned yesterday I was running off a IOzone benchmark on EC2 to see how the disk performs, after reading about it in this online benchmark article.
There are a couple of nice features with this benchmark
- Output is saved in a format ready for surface area graphs in Excel
- Gnuplot options available as well
- It tests stride size to see if there are any stripe boundary or IO library issues.
Another first is you can download my results in this file iozone_benchmark_ec2.zip served from Amazon S3 (Right click SAVE AS)
The IOzone documentation[PDF] which is short, but detailed.
- wget http://www.iozone.org/src/current/iozone-3-283.i386.rpm
- rpm -Uvh iozone-3-283.i386.rpm
- export PATH=$PATH:/opt/iozone/bin
Running a benchmark:
Note: File should be larger than available memory. -g 2G indicates 2 Gigabyte file.
- iozone -Ra -g 2G > iozone.out
The full size graphs can be found via Amazon S3
hi,>can you tell me how to generate the graphs after running iozone?
Grab the zip file I posted>< HREF="http://s3.amazonaws.com/dbadojo_benchmark/iozone_benchmark_ec2.zip" REL="nofollow">http://s3.amazonaws.com/dbadojo_benchmark/iozone_benchmark_ec2.zip<>>>I generated the 3D graphs and the surface area (heatmap) graphs in excel.>>To generate either, you will need to cut and paste the IO Zone output (it will output in csv format) into excel.>Not sure OpenOffice has surface area graphs, however if you google around you might find an answer.
I dont really understand how -g option works with IOZone.
Its says ” Maximum File size, Generally double the size of your RAM”
Cud you please help me on this
The -g option is the size of the file which you want iozone to use.
The reason for using a file larger than the amount RAM (physical memory) is to force the OS to read from disk rather than from memory.
Most OS (Win/Unix/Linux/OSX) will cache a file in memory if it will fit, as it is faster.
By making the file larger than memory it forces the OS to read the disk which is what you are benchmarking.