Difference between revisions of "Best practices on Lustre parallel file systems"

From ScientificComputing
Jump to: navigation, search
(Optimizing HDF5 file layout for best performance on Lustre)
 
(20 intermediate revisions by 2 users not shown)
Line 1: Line 1:
<!--
 
 
==Introduction==
 
==Introduction==
Lustre is a type of parallel distributed file system, generally used for large-scale cluster computing. Files are distributed across multiple servers, and then striped across multiple disks.
+
On the Euler and the Leonhard cluster, the global '''scratch''' and '''work''' directories
  
The Lustre file system is shared among many users. It is optimized for parallel I/O and large files. Please note that
+
/cluster/scratch/$USER
 +
/cluster/work/
 +
 
 +
are hosted on Lustre file systems. The are optimized especially for '''parallel I/O''' and '''large files'''. Those file systems are '''shared among many users'''. If you are
  
 
*working with a large number of small files
 
*working with a large number of small files
Line 9: Line 11:
 
*accessing the same file with hundreds of processes
 
*accessing the same file with hundreds of processes
  
will not only slow down your jobs. It can overload the entire file system affecting all users. Therefore please carefully read our best practices guide before using <tt>/cluster/work</tt> or <tt>/cluster/scratch</tt>.
+
then this will not only slow down your jobs. '''It can overload the entire file system affecting all users'''. Therefore please carefully read our [[#Best_practices|best practices guide]] before using <tt>/cluster/work</tt> and/or <tt>/cluster/scratch</tt>.
  
 
==Lustre architecture==
 
==Lustre architecture==
 +
Lustre is a '''parallel distributed file system'''. Files are distributed across multiple servers, and then striped across multiple disks.
 +
 
A Lustre file system has three major functional units:
 
A Lustre file system has three major functional units:
  
* Metadata servers (MDS) that stores namespace metadata, such as filenames, directories, access permissions, and file layout.  
+
* '''Metadata servers (MDS)''' that stores namespace metadata, such as filenames, directories, access permissions, and file layout.  
* Object storage server (OSS) nodes that store file data on one or more object storage target (OST) devices.
+
* '''Object storage server (OSS)''' nodes that store file data on one or more object storage target (OST) devices.
* Client(s) that access and use the data.
+
* '''Client(s)''' that access and use the data.
  
When a client accesses a file, it performs a filename lookup on the MDS. When the MDS filename lookup is complete and the user and client have permission to access and/or create the file, then the layout of an existing file is returned a new file is created.
+
When a client accesses a file, it performs a filename lookup on the MDS. When the MDS filename lookup is complete and the user and client have permission to access the file, then the layout of the file is returned.
  
 
For read or write operations, the client then interprets the file layout, which maps the file logical offset and size to one or more objects, each residing on a separate OST. The client then locks the file range being operated on and executes one or more parallel read or write operations directly to the OSS nodes.
 
For read or write operations, the client then interprets the file layout, which maps the file logical offset and size to one or more objects, each residing on a separate OST. The client then locks the file range being operated on and executes one or more parallel read or write operations directly to the OSS nodes.
Line 25: Line 29:
  
 
==Best practices==
 
==Best practices==
 +
===Avoid unnecessary I/O operations===
 +
In many programs, there are options to control I/O to make them more or less verbose. In general I/O operations are slowing down your computation, because during I/O operations the CPU is waiting and doing nothing. Therefore only do I/O if it is required and provides an added value to your computation. Otherwise try to avoid unneccessary I/O operations whenever possible.
 +
 
===Limit repetitive Open/Close operations===
 
===Limit repetitive Open/Close operations===
 
If you need to write a lot of values into a file as part of a loop, then there are multiple ways of achieving this task. Please make sure that you never put the open and close statements inside the loop as shown in this Python example:
 
If you need to write a lot of values into a file as part of a loop, then there are multiple ways of achieving this task. Please make sure that you never put the open and close statements inside the loop as shown in this Python example:
Line 41: Line 48:
  
 
===Limit repetitive "stat" operations===
 
===Limit repetitive "stat" operations===
If you are running a code that needs at some point to check if a file exists, it is sufficient to check for this every few seconds
+
If you are running a code that needs at some point to wait until a certain file is created, it is sufficient to check for this every 5-10 seconds. Checking this without any delay in a loop can cause more than 1000 checks per second create a lot of unnecessary file stat calls.
  
===Directory listings: ls vs. ls -l===
+
===Directory listings: ls vs. ls -l vs. ls --color===
If you run the <tt>ls</tt> command for listing a file or a directory, then it will query the MDS for this information. But when running the command with the <tt>-l</tt> option, it will also need to access the OSS to look up the file size, which creates additional load on the storage system.
+
If you run the <tt>ls</tt> command for listing a file or a directory, then it will query the MDS for this information. But when running the command with the <tt>-l</tt> option, it will also need to access the OSSes to look up the file size, which creates additional load on the storage system.
  
 
* Use <tt>ls</tt> if you would like to list files and directories
 
* Use <tt>ls</tt> if you would like to list files and directories
 
* Only use <tt>ls -l</tt> if you also need to know about the file size
 
* Only use <tt>ls -l</tt> if you also need to know about the file size
 +
* Unset the default <tt>color</tt> option for <tt>ls</tt> (<tt>ls --color</tt>) if you need to do ls on Lustre directories with many files: <tt>ls --color=never</tt> or to permanently disable it during your session <tt>$ \ls</tt>
  
===Don't store a large number of files in a single directory===
+
===Use subdirectories instead of storing all files in a single directory===
 +
When a file is accessed, Lustre puts a lock on the parent directory. If many files are opened in the same directory, then this will cause contention. To minimize contention, distribute your files into a subdirectory structure. This way your files are also organized and easier to handle.
  
===Avoid Accessing Small Files on Lustre Filesystems===
+
If you accidentally run an <tt>ls -l</tt> instead of just <tt>ls</tt>, then it also makes a difference if the directory contains 20 or 100'000 files.
  
===Use a Stripe Count of 1 for Directories with Many Small Files===
+
===Try to use local scratch for serial jobs===
 +
If your data set fits into local scratch (≤ 300 GB), then try to use local scratch instead of Lustre as it will be faster in almost all cases.
  
===Avoid Accessing Executables on Lustre Filesystems===
+
https://scicomp.ethz.ch/wiki/Using_local_scratch
  
===Increase the Stripe Count for Parallel Access to the Same File===
+
A typical workflow could be to copy your files from Lustre to local scratch at the beginning of a job, then process the files and copy back the results of the job from local scratch to Lustre.
  
===Restripe Large Files===
+
Jobs working with many small files might be improved by a huge factor with the latency reduction provided by local scratch.
  
===Limit the Number of Processes Performing Parallel I/O===
+
===Use other storage locations for small files===
 +
The Lustre file system is the worst place to store a lot of small files. Other file systems like $HOME or local scratch ($TMPDIR, only on compute nodes) are much better suited to deal with small files. If you have to store a lot of small files on Lustre, then please at least tar them up to a single file. For processing those files, untar them to local scratch at the beginning of the job, process them on the compute node and at the end of the job tar up the results and copy back the archive to Lustre.
  
===Avoid Having Multiple Processes Open the Same File(s) at the Same Time===
+
===Avoid reading the same region of a file from many processes at the same time===
 
+
If you are running a series of jobs in parallel that are all accessing the same region of a file, then this will cause performance problems. It might be better to either split the file into parts that can be used by individual jobs, or to work on multiple copies of the same file. Each job could for instance copy the file to local scratch in order to avoid contention.
==Troubleshooting==
 
  
 
==Working with stripes (advanced users)==
 
==Working with stripes (advanced users)==
Line 71: Line 81:
  
 
===How to display the current striping settings===
 
===How to display the current striping settings===
The default stripe setting of a file or directory can be shown with the command '''lfs getstripe''':
+
The default stripe setting of a file or directory can be shown with the command '''lfs getstripe'''. It is configured by default to 1:
  
 
  [sfux@eu-login-24-ng ~]$ lfs getstripe $SCRATCH/__USAGE_RULES__  
 
  [sfux@eu-login-24-ng ~]$ lfs getstripe $SCRATCH/__USAGE_RULES__  
Line 85: Line 95:
 
  [sfux@eu-login-24-ng ~]$
 
  [sfux@eu-login-24-ng ~]$
  
For directories, use the -d option
+
* stripe_count = 1 : Use the filesystem default stripe count
 +
* stripe_size = 1048576 : Use 1 MiB stripe/chunk size
 +
* stripe_offset = -1: Let Lustre choose the next OST (you shouldn't change this)
 +
 
 +
===Hints for proper striping count===
 +
 
 +
The best stripping count depends mostly on the IO pattern access, the file size and the number of nodes used to access a file. In order to help users distribute their data this command will automatically set the proper striping depending on the size of the file. The striping will adapt to the size of the file while the file grows:
  
[sfux@eu-login-24-ng ~]$ lfs getstripe -d $SCRATCH
 
stripe_count:  1 stripe_size:    1048576 stripe_offset:  -1
 
[sfux@eu-login-24-ng ~]$
 
  
* stripe_count = -1 : Use the filesystem default stripe count (= spread data to all OSTs)
+
lfs setstripe -E 500M -c 1 -E 10G -c 2 -E 50G -c 4 -E 200G -c 8 -E -1 -c 20 <file_to_be_created>
* stripe_size = 1048576 : Use 1 MiB stripe/chunk size
 
* stripe_offset = -1: Let Lustre choose the next OST (you shouldn't change this)
 
  
 
===How to change stripe settings===
 
===How to change stripe settings===
The stripe setting of a directory can be changed with the command '''lfs setstripe'''.
+
The stripe setting of a directory can be changed with the command '''lfs setstripe''':
  
Note!
+
[sfux@eu-login-02-ng test]$ lfs setstripe -E 500M -c 1 -E 10G -c 2 -E 50G -c 4 -E 200G -c 8 -E -1 -c 20 my_directory
 +
 
 +
Please note:
 
* You '''can not''' change the striping of '''existing files'''
 
* You '''can not''' change the striping of '''existing files'''
* You '''can''' always change the striping parameters of an existing directory
+
* You '''can''' always change the striping parameters for new files with the <tt>lfs command</tt>
* It is possible to create files with non-default striping parameters with the lfs command
 
 
* A subdirectory '''inherits''' all stripe parameters from its parent directory (if not changed via lfs setstripe)
 
* A subdirectory '''inherits''' all stripe parameters from its parent directory (if not changed via lfs setstripe)
  
-->
+
Example:
 +
 
 +
[sfux@eu-login-02-ng test]$ lfs setstripe -E 500M -c 1 -E 10G -c 2 -E 50G -c 4 -E 200G -c 8 -E -1 -c 20 my_new_file
 +
 
 +
The file will be precreated with zero bytes
 +
 
 +
Change the stripe size of a file:
 +
 
 +
* Create and empty file with the preferred stripe settings
 +
* Copy the original file into the new one
 +
* Move the new one to the original file name
 +
 
 +
Example:
 +
 
 +
[sfux@eu-login-02-ng test]$ pwd
 +
/cluster/scratch/sfux/test
 +
[sfux@eu-login-02-ng test]$ ls
 +
DFT.tar.gz
 +
[sfux@eu-login-02-ng test]$ lfs getstripe DFT.tar.gz
 +
DFT.tar.gz
 +
lmm_stripe_count:  1
 +
lmm_stripe_size:    1048576
 +
lmm_pattern:        1
 +
lmm_layout_gen:    0
 +
lmm_stripe_offset:  3
 +
        obdidx          objid          objid          group
 +
              3      218244665      0xd022639                0
 +
 
 +
The original file has a stripe count of 1. Now it is changed to the maximal stripe count (-1 sets it to the maximal stripe count):
 +
 
 +
[sfux@eu-login-02-ng test]$ lfs setstripe -E 500M -c 1 -E 10G -c 2 -E 50G -c 4 -E 200G -c 8 -E -1 -c 20 DFT.tar.gz_tmp
 +
[sfux@eu-login-02-ng test]$ cp -a DFT.tar.gz DFT.tar.gz_tmp
 +
[sfux@eu-login-02-ng test]$ mv DFT.tar.gz_tmp DFT.tar.gz
 +
[sfux@eu-login-02-ng test]$
 +
 
 +
==Optimizing HDF5 file layout for best performance on Lustre==
 +
For HDF5 files, there are different ways to create the file layout when storing data. There are some guides about how one can optimize HDF5 function calls and the layout of the HDF5 file to get optimal performance on Lustre file systems:
 +
 
 +
* [[media:Tuning_HDF5_for_Lustre_File_Systems.pdf|Tuning HDF5 for Lustre file systems]]
 +
* [https://www.nersc.gov/users/training/online-tutorials/introduction-to-scientific-i-o/?start=5 Introduction to scientific I/O from NERSC]

Latest revision as of 10:09, 11 March 2022

Introduction

On the Euler and the Leonhard cluster, the global scratch and work directories

/cluster/scratch/$USER
/cluster/work/

are hosted on Lustre file systems. The are optimized especially for parallel I/O and large files. Those file systems are shared among many users. If you are

  • working with a large number of small files
  • running thousands of unnecessary I/O operations per second (running Open/Close in a loop)
  • accessing the same file with hundreds of processes

then this will not only slow down your jobs. It can overload the entire file system affecting all users. Therefore please carefully read our best practices guide before using /cluster/work and/or /cluster/scratch.

Lustre architecture

Lustre is a parallel distributed file system. Files are distributed across multiple servers, and then striped across multiple disks.

A Lustre file system has three major functional units:

  • Metadata servers (MDS) that stores namespace metadata, such as filenames, directories, access permissions, and file layout.
  • Object storage server (OSS) nodes that store file data on one or more object storage target (OST) devices.
  • Client(s) that access and use the data.

When a client accesses a file, it performs a filename lookup on the MDS. When the MDS filename lookup is complete and the user and client have permission to access the file, then the layout of the file is returned.

For read or write operations, the client then interprets the file layout, which maps the file logical offset and size to one or more objects, each residing on a separate OST. The client then locks the file range being operated on and executes one or more parallel read or write operations directly to the OSS nodes.

After the initial lookup of the file layout, the MDS is not normally involved in file IO operations since all block allocation and data IO is managed internally by the OST. Clients do not directly modify the objects or data on the OST filesystems, but instead delegate this task to OSS nodes.

Best practices

Avoid unnecessary I/O operations

In many programs, there are options to control I/O to make them more or less verbose. In general I/O operations are slowing down your computation, because during I/O operations the CPU is waiting and doing nothing. Therefore only do I/O if it is required and provides an added value to your computation. Otherwise try to avoid unneccessary I/O operations whenever possible.

Limit repetitive Open/Close operations

If you need to write a lot of values into a file as part of a loop, then there are multiple ways of achieving this task. Please make sure that you never put the open and close statements inside the loop as shown in this Python example:

for i in range(1000):
    f=open('test2.txt', 'a')
    f.write(some_data)
    f.close()

This will cause that the same file is opened and closed 1000 times, which causes a total of 2000 I/O operations and 1998 of them are unnecessary. It is sufficient to open the file once, then write all values to it and close it at the end, resulting in only 2 I/O operations

f=open('test1.txt', 'w')
for i in range(1000):
    f.write(some_data)
f.close()

Limit repetitive "stat" operations

If you are running a code that needs at some point to wait until a certain file is created, it is sufficient to check for this every 5-10 seconds. Checking this without any delay in a loop can cause more than 1000 checks per second create a lot of unnecessary file stat calls.

Directory listings: ls vs. ls -l vs. ls --color

If you run the ls command for listing a file or a directory, then it will query the MDS for this information. But when running the command with the -l option, it will also need to access the OSSes to look up the file size, which creates additional load on the storage system.

  • Use ls if you would like to list files and directories
  • Only use ls -l if you also need to know about the file size
  • Unset the default color option for ls (ls --color) if you need to do ls on Lustre directories with many files: ls --color=never or to permanently disable it during your session $ \ls

Use subdirectories instead of storing all files in a single directory

When a file is accessed, Lustre puts a lock on the parent directory. If many files are opened in the same directory, then this will cause contention. To minimize contention, distribute your files into a subdirectory structure. This way your files are also organized and easier to handle.

If you accidentally run an ls -l instead of just ls, then it also makes a difference if the directory contains 20 or 100'000 files.

Try to use local scratch for serial jobs

If your data set fits into local scratch (≤ 300 GB), then try to use local scratch instead of Lustre as it will be faster in almost all cases.

https://scicomp.ethz.ch/wiki/Using_local_scratch

A typical workflow could be to copy your files from Lustre to local scratch at the beginning of a job, then process the files and copy back the results of the job from local scratch to Lustre.

Jobs working with many small files might be improved by a huge factor with the latency reduction provided by local scratch.

Use other storage locations for small files

The Lustre file system is the worst place to store a lot of small files. Other file systems like $HOME or local scratch ($TMPDIR, only on compute nodes) are much better suited to deal with small files. If you have to store a lot of small files on Lustre, then please at least tar them up to a single file. For processing those files, untar them to local scratch at the beginning of the job, process them on the compute node and at the end of the job tar up the results and copy back the archive to Lustre.

Avoid reading the same region of a file from many processes at the same time

If you are running a series of jobs in parallel that are all accessing the same region of a file, then this will cause performance problems. It might be better to either split the file into parts that can be used by individual jobs, or to work on multiple copies of the same file. Each job could for instance copy the file to local scratch in order to avoid contention.

Working with stripes (advanced users)

Lustre will always try to distribute your data across all OSTs. The striping parameters can be tuned per file or directory.

How to display the current striping settings

The default stripe setting of a file or directory can be shown with the command lfs getstripe. It is configured by default to 1:

[sfux@eu-login-24-ng ~]$ lfs getstripe $SCRATCH/__USAGE_RULES__ 
/cluster/scratch/sfux/__USAGE_RULES__
lmm_stripe_count:   1
lmm_stripe_size:    1048576
lmm_pattern:        1
lmm_layout_gen:     0
lmm_stripe_offset:  3
        obdidx           objid           objid           group
             3          619261        0x972fd                0 

[sfux@eu-login-24-ng ~]$
  • stripe_count = 1 : Use the filesystem default stripe count
  • stripe_size = 1048576 : Use 1 MiB stripe/chunk size
  • stripe_offset = -1: Let Lustre choose the next OST (you shouldn't change this)

Hints for proper striping count

The best stripping count depends mostly on the IO pattern access, the file size and the number of nodes used to access a file. In order to help users distribute their data this command will automatically set the proper striping depending on the size of the file. The striping will adapt to the size of the file while the file grows:


lfs setstripe -E 500M -c 1 -E 10G -c 2 -E 50G -c 4 -E 200G -c 8 -E -1 -c 20 <file_to_be_created>

How to change stripe settings

The stripe setting of a directory can be changed with the command lfs setstripe:

[sfux@eu-login-02-ng test]$ lfs setstripe -E 500M -c 1 -E 10G -c 2 -E 50G -c 4 -E 200G -c 8 -E -1 -c 20 my_directory

Please note:

  • You can not change the striping of existing files
  • You can always change the striping parameters for new files with the lfs command
  • A subdirectory inherits all stripe parameters from its parent directory (if not changed via lfs setstripe)

Example:

[sfux@eu-login-02-ng test]$ lfs setstripe -E 500M -c 1 -E 10G -c 2 -E 50G -c 4 -E 200G -c 8 -E -1 -c 20 my_new_file

The file will be precreated with zero bytes

Change the stripe size of a file:

  • Create and empty file with the preferred stripe settings
  • Copy the original file into the new one
  • Move the new one to the original file name

Example:

[sfux@eu-login-02-ng test]$ pwd
/cluster/scratch/sfux/test
[sfux@eu-login-02-ng test]$ ls
DFT.tar.gz
[sfux@eu-login-02-ng test]$ lfs getstripe DFT.tar.gz 
DFT.tar.gz
lmm_stripe_count:   1
lmm_stripe_size:    1048576
lmm_pattern:        1
lmm_layout_gen:     0
lmm_stripe_offset:  3
        obdidx           objid           objid           group
             3       218244665      0xd022639                0

The original file has a stripe count of 1. Now it is changed to the maximal stripe count (-1 sets it to the maximal stripe count):

[sfux@eu-login-02-ng test]$ lfs setstripe -E 500M -c 1 -E 10G -c 2 -E 50G -c 4 -E 200G -c 8 -E -1 -c 20 DFT.tar.gz_tmp
[sfux@eu-login-02-ng test]$ cp -a DFT.tar.gz DFT.tar.gz_tmp
[sfux@eu-login-02-ng test]$ mv DFT.tar.gz_tmp DFT.tar.gz
[sfux@eu-login-02-ng test]$

Optimizing HDF5 file layout for best performance on Lustre

For HDF5 files, there are different ways to create the file layout when storing data. There are some guides about how one can optimize HDF5 function calls and the layout of the HDF5 file to get optimal performance on Lustre file systems: