Difference between revisions of "How to use MDSS tape storage at NCI"

m
Line 1: Line 1:
 +
 
Massdata (Mass Data Storage System, MDSS for short) is the tape storage available at NCI. This kind of storage is intended for long term archiving of large files. Each project has a directory on the MDSS, the amount of storage allocated depends on the project allocation and can be checked using the nci_account command.
 
Massdata (Mass Data Storage System, MDSS for short) is the tape storage available at NCI. This kind of storage is intended for long term archiving of large files. Each project has a directory on the MDSS, the amount of storage allocated depends on the project allocation and can be checked using the nci_account command.
  
There is a comprehensive NCI guide to using MDSS
+
=== '''MDSS proper usage''' ===
 +
 
 +
MDSS is designed for medium to long-term archive of large files, so it is suitable for
 +
 
 +
*Files you are required to keep, for example model outputs or configurations from published datasets, publications, PhD thesis etc.
 +
*Files that you or someone else are likely to reuse or analyse again in the future but not in the next few months. For example restart files or other model output you are not immediately using should be moved from disk to mdss as soon as possible.
 +
*MDSS is suitable to backup  big data projects, like model output which could not be backed up elsewhere. It is not suitable for small amounts of data where using other backup options would be easier and more efficient. It is also not suitable for code files you might want to keep, for this you online services as Github or Bitbucket should be your preferred choice.
 +
 
 +
=== '''Guidelines for storage''' ===
 +
 
 +
*Big files: if your files are small in size (less than 20Mb) then use tools like tar to bundle them into a single archive file
 +
*Files should be group readable, with group execute permissions for directories. This helps with long term maintenance, allowing administrators to track the type and size of archived data.
 +
 
 +
 
 +
 
 +
=== Accessing MDSS ===
 +
 
 +
Massdata cannot be accessed directly via a directory path. All access of MDSS is via the command '''mdss.'''
 +
 
 +
Users connected to the project have rwx permissions in that directory and so may create their own files in those areas.
 +
 
 +
Mdss has several sub-commands and options to see all of them:
  
===MDSS proper usage===
+
<syntaxhighlight lang="bash">
 +
mdss --help or
  
MDSS is designed for medium to long-term archive of large files, so it is suitable for
+
man mdss;
 +
</syntaxhighlight>
  
* Files you are required to keep, for example model outputs or configurations from published datasets, publications, PhD thesis etc.
+
Usually you specify the project, if you don't it will use your default project, and then add a sub-command and the path of the files and directories you want to upload, list etc.
* Files that you or someone else are likely to reuse or analyse again in the future but not in the next few months. For example restart files or other model output you are not immediately using should be moved from disk to mdss as soon as possible.
 
  
===Guidelines for storage===
+
mdss -P <project-id>;+ <sub-command> + <path>
 +
Most useful sub-commands are:
  
* Big files: if your files are small in size (less than 20Mb) then use tools like tar to bundle them into a single archive file
+
<syntaxhighlight lang="bash">
* Files should be group readable, with group execute permissions for directories. This helps with long term maintenance, allowing administrators to track the type and size of archived data.
+
*mdss put&nbsp;- to upload files  
 +
*mdss get&nbsp;- to retrieve files  
 +
*mdss ls &nbsp;- to list directories and files
 +
*mdss dmdu&nbsp;&nbsp;- to get the size of a directory
  
MDSS is not a backup service and it is not suitable for code files you might want to keep, for this you might prefer to use online services as Github or Bitbucket.
+
NB &nbsp;"mdss du" will also work but only return the size of what is still cached, dmdu will give the full size of what is on tape regardless that is cached or not.
  
===Accessing MDSS===
+
*mdss dmls - to see what is currently online in the cache, what is on tape
Massdata cannot be accessed directly via a directory path.
+
<syntaxhighlight>
All access of MDSS is via the command mdss
 
mdss -P project-id + command (ls, put, get ..)
 
See mdss -help to get a full list of the subcommands and
 
Mdss --help subcommand to get specific help
 
 
Please note mdss commands work only interactively or with ‘copyq’
 
Please note mdss commands work only interactively or with ‘copyq’
* Users connected to the project have rwx permissions in that directory and so may create their own files in those areas.
 
 
  
 
=== Preparing your data for mdss ===
 
=== Preparing your data for mdss ===
  
#Organise your files and delete anything which you won’t be re-using. It is tempting to copy entire directories as they are to mdss thinking you’ll be getting back to them again later. There is currently no easy way to list what you are storing on massdata and so trying to tidy up after you uploaded your files would be slow and painful. Even more than with other storage options, it is really important to put there only suitable files and make sure that they have been compressed and tarred together if necessary  
+
#Organise your files and delete anything which you will not&nbsp;be re-using. It is tempting to copy entire directories as they are thinking you will get&nbsp;back to them again later. There is currently no easy way to list what you are storing on massdata and so trying to tidy up after you uploaded your files would be slow and painful. Even more than with other storage options, it is really important to put there only suitable files and make sure that they have been compressed and tarred together if necessary.
 
#NCI guidelines suggest a minimum size of 20MB per file and an average size of 250MB.  
 
#NCI guidelines suggest a minimum size of 20MB per file and an average size of 250MB.  
  
Line 40: Line 61:
 
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp;[[NetCDF_Compression_Tools|Compressing tools]]
 
&nbsp; &nbsp; &nbsp; &nbsp; &nbsp;[[NetCDF_Compression_Tools|Compressing tools]]
  
===Monitoring mdss usage===  
+
=== Monitoring mdss usage ===
  
nci_account -P <project-id> will give an account of the total massdata allocation, usage and availability both as size and i-nodes (i.e. number of files and directories)
+
In the past you could nci_account to monitor the allocation and how much of it was still available. Currently this is not possible anymore so particularly if you want to move a big amount of data to mdss you should first check with the lead CI of the project you wan tto use, to make sure enough storage is available.
<syntaxhighlight lang=bash>
+
Unfortunately, there is also not a command to check quickly usage by user-id as for /g/data and /short. The only way to get this information currently is to ask help@nci.org.au, administrators can access this information for any CI of the group.  
Usage Report: Project=w35 Storage Period=2017.9 (01/07/2017-30/09/2017)
+
=== Transferring data to and from MDSS ===
=======================================================================
 
-------------------------------------------------------------------------------------------------
 
System    StoragePt            Grant      Usage      Avail      iGrant      iUsage      iAvail
 
-------------------------------------------------------------------------------------------------
 
dmf      massdata          4048.00GB  1209.00GB  2839.00GB    323.00K      8.88K    314.12K
 
global    gdata1              76.00TB    59.69TB    16.31TB    3883.00K    2299.01K    1583.99K
 
global    gdata1a            76.00TB    60.47TB    15.53TB    3883.00K    2313.58K    1569.42K
 
raijin    short              15.00TB      7.54TB      7.46TB    3280.00K    2767.07K    512.93K
 
-------------------------------------------------------------------------------------------------
 
Total                        170.95TB    128.88TB    42.08TB      11.37M      7.39M      3.98M
 
</syntaxhighlight>
 
Unfortunately there is not a command to check quickly usage by user-id as for /g/data and /short.
 
The only way to get this information currently is to ask help@nci.org.au, administrators can access this information for any CI of the group.
 
  
===Transferring data to and from MDSS===
 
 
NCI supports different commands to work with MDSS as it is explained on their User Guide. The CMS team has also developed a utility called mdssdiff. This utility allows users to compare the contents of the local directory and a directory under /massdata. It will also recursively update the content on the massdata directory to copy the local directory or vice versa.
 
NCI supports different commands to work with MDSS as it is explained on their User Guide. The CMS team has also developed a utility called mdssdiff. This utility allows users to compare the contents of the local directory and a directory under /massdata. It will also recursively update the content on the massdata directory to copy the local directory or vice versa.
  
===Modifications to MDSS datasets===  
+
=== Modifications to MDSS datasets ===
 +
 
 
Contact NCI at help@nci.org.au if large metadata operations are needed on massdata, as changing ownership, project code, permissions etc. of existing datasets
 
Contact NCI at help@nci.org.au if large metadata operations are needed on massdata, as changing ownership, project code, permissions etc. of existing datasets
  
 
[[Category:NCI Guidelines]]
 
[[Category:NCI Guidelines]]

Revision as of 23:19, 15 July 2021

Massdata (Mass Data Storage System, MDSS for short) is the tape storage available at NCI. This kind of storage is intended for long term archiving of large files. Each project has a directory on the MDSS, the amount of storage allocated depends on the project allocation and can be checked using the nci_account command.

MDSS proper usage

MDSS is designed for medium to long-term archive of large files, so it is suitable for

  • Files you are required to keep, for example model outputs or configurations from published datasets, publications, PhD thesis etc.
  • Files that you or someone else are likely to reuse or analyse again in the future but not in the next few months. For example restart files or other model output you are not immediately using should be moved from disk to mdss as soon as possible.
  • MDSS is suitable to backup  big data projects, like model output which could not be backed up elsewhere. It is not suitable for small amounts of data where using other backup options would be easier and more efficient. It is also not suitable for code files you might want to keep, for this you online services as Github or Bitbucket should be your preferred choice.

Guidelines for storage

  • Big files: if your files are small in size (less than 20Mb) then use tools like tar to bundle them into a single archive file
  • Files should be group readable, with group execute permissions for directories. This helps with long term maintenance, allowing administrators to track the type and size of archived data.

 

Accessing MDSS

Massdata cannot be accessed directly via a directory path. All access of MDSS is via the command mdss.

Users connected to the project have rwx permissions in that directory and so may create their own files in those areas.

Mdss has several sub-commands and options to see all of them:

mdss --help or

man mdss;

Usually you specify the project, if you don't it will use your default project, and then add a sub-command and the path of the files and directories you want to upload, list etc.

mdss -P <project-id>;+ <sub-command> + <path> Most useful sub-commands are:

<syntaxhighlight lang="bash">

  • mdss put - to upload files
  • mdss get - to retrieve files
  • mdss ls  - to list directories and files
  • mdss dmdu  - to get the size of a directory

NB  "mdss du" will also work but only return the size of what is still cached, dmdu will give the full size of what is on tape regardless that is cached or not.

  • mdss dmls - to see what is currently online in the cache, what is on tape

<syntaxhighlight> Please note mdss commands work only interactively or with ‘copyq’

Preparing your data for mdss

  1. Organise your files and delete anything which you will not be re-using. It is tempting to copy entire directories as they are thinking you will get back to them again later. There is currently no easy way to list what you are storing on massdata and so trying to tidy up after you uploaded your files would be slow and painful. Even more than with other storage options, it is really important to put there only suitable files and make sure that they have been compressed and tarred together if necessary.
  2. NCI guidelines suggest a minimum size of 20MB per file and an average size of 250MB.

While you are preparing your data to be moved it is an opportunity to also document, if you haven’t done so already, what you are archiving and how. Even a simple readme file added to your main directory can help others and your future self. If you are archiving data underlying a publication or published dataset then it is important a summary of what is stored in /massdata and how is part of the dataset management plan.

Useful tools:

         TAR - to create archives

         Compressing tools

Monitoring mdss usage

In the past you could nci_account to monitor the allocation and how much of it was still available. Currently this is not possible anymore so particularly if you want to move a big amount of data to mdss you should first check with the lead CI of the project you wan tto use, to make sure enough storage is available. Unfortunately, there is also not a command to check quickly usage by user-id as for /g/data and /short. The only way to get this information currently is to ask help@nci.org.au, administrators can access this information for any CI of the group.

Transferring data to and from MDSS

NCI supports different commands to work with MDSS as it is explained on their User Guide. The CMS team has also developed a utility called mdssdiff. This utility allows users to compare the contents of the local directory and a directory under /massdata. It will also recursively update the content on the massdata directory to copy the local directory or vice versa.

Modifications to MDSS datasets

Contact NCI at help@nci.org.au if large metadata operations are needed on massdata, as changing ownership, project code, permissions etc. of existing datasets