Performance tuning tips
To get best throughput on the scratch file system (/cluster/work), you may need to change the data striping. Striping shall be adjusted based on the client access pattern to optimally load the object storage targets (OSTs). On Lustre, the OSTs are referring to disks or storage volumes constructing the whole file system.
stripe_count indicates how many OSTs to use.
stripe_size indicates how much data to write to one OST before moving to
the next OST.
- striping will only take affect only on new files, created or copied into the specified directory or file name;
- default stripe_count on /cluster is 1.
For more detailed information on striping, please consult the Lustre documentation.
Check out current striping
To see current stripe size, use
lfs getsripe [file_system, dir, file]
lfs getstripe /cluster/tmp/test /cluster/tmp/test stripe_count: 1 stripe_size: 1048576 stripe_offset: -1
Note: stripe size is shown in bytes.
For large files it is advisable to increase stripe count and perhaps chunk size too. e.g.:
# stripe huge file across 8 OSTs lfs setstripe --stripe-count 8 "my_file" # stripe across 4 OSTs using 8MB chunks. lfs setstripe --stripe-size 8M --stripe-count 4 "my_dir"
It is advisable to use higher stripe count for scientific application that writes to a single file from hundreds of nodes, or a binary executable that is loaded by many nodes when an application starts.
Choose a stripe size between 1MB and 4MB for sequential I/O. Larger than 4MB stripe size may result in performance loss in case of shared files.
Set the stripe size a multiple of the write() size, if your application is writing in a consistent and aligned way.
For many small files and one client accessing each file, change stripe count to 1. Avoid having small files with large stripe counts. This negatively impacts the performance due to the unnecessary communication to multiple OSTs.
lfs setstripe --stripe-count 1 "my_dir"
- Avoid having a large number of files in a single directory and rather split files in multiple sub-directories.
- Avoid accessing small files. Use
/node/scratchwhenever possible instead of
/cluster/workfor small files.
- Avoid repetitive
statoperations because it creates a significant load on the file system.
- Do not use
ls -lon whole directory because it issues RPCs to the MDS for each file and directory it lists. Use rather
ls -lonly for the specific file you need extended information such as: permissions, ownership, etc.