Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision
Previous revision
doku:vsc3_storage [2017/06/02 11:36]
sreinwal
doku:vsc3_storage [2021/08/23 08:52] (current)
goldenberg [Quotas]
Line 1: Line 1:
-~~NOTOC~~ 
  
 ===== VSC-3 Storage ===== ===== VSC-3 Storage =====
  
-VSC-3 provides three facilities for persisting data: the high-performance BeeGFS Parallel Filesystem (former Fraunhofer Parallel Filesystem, FhGFS)the Network File System (NFSand a node-local ramdisk. They are accessible under:+**The ''$GLOBAL'' (and ''$SCRATCH''system has been decommissioned. Use ''$BINFL'' or ''$DATA'' (if the project also exists on VSC4)**
  
-  * **NFS**: $HOME which expands to /home/lv<project>/<username> +This article is about the ''$GLOBAL'' and ''$HOME'' filesystems of VSC-3. If you are searching for info about the bioinformatics storagethe article can be found [[binf_nodes|here]].
-  * **BeeGFS (former FhGFS)**: $GLOBAL expands to /fhgfs/global/lv<project>/<username>$SCRATCH to /fhgfs/<node> +
-  * **Scratch RAM Disk** $TMPDIR </tmp>+
  
 +VSC-3 provides three facilities for persisting data: the high-performance BeeGFS Parallel Filesystem (former Fraunhofer Parallel Filesystem, FhGFS), the Network File System (NFS) and a node-local ramdisk. They are accessible under:
  
-==== Usage $HOME ====+  * **NFS**: ''$HOME'' which expands to /home/lv<project>/<username> 
 +  * **BeeGFS (former FhGFS)**: **Decommissioned** 
 +     * ''$GLOBAL'' expands to /fhgfs/global/lv<project>/<username>,  
 +     * ''$SCRATCH'' to /fhgfs/<node> (node local) 
 +  * **Scratch RAM Disk** ''$TMPDIR'' </tmp>
  
-$HOME is the location of the user UNIX home directory. It can be accessed from login and compute nodes. $HOME can be used to hold results, settings, source code etc. - data for which high concurrent job throughput and support for large file sizes is not required. Conversely, the parallel BeeGFS filesystem (see below) should utilized to persist temporary data in compute runs. 
  
-Backup of $HOME is user responsibility. +==== Usage ''$HOME'' ====
  
-$HOME is provided from file servers with disk arrays that are exported over the network file system (NFS)Even on highly scaled storage such on VSC-3, the number of concurrent file operations is bound by spinning disk physics: small file (writeoperations can easily saturate capacity. Hence, please mind that $HOME is a shared resource over all projects on a given NFS server. In case your project requires persistence over a large number of small files please contact VSC administration in advance   +''$HOME'' is the location of the user UNIX home directory. It can be accessed from login and compute nodes. $HOME can be used to hold results, settings, source code etc. - data for which high concurrent job throughput and support for large file sizes is not required. Conversely, the parallel BeeGFS filesystem (see belowshould utilized to persist temporary data in compute runs.
  
 +Backup of ''$HOME'' is user responsibility. 
  
-==== Scratch Space Usage: $GLOBAL and $SCRATCH ====+''$HOME'' is provided from file servers with disk arrays that are exported over the network file system (NFS). Even on highly scaled storage such on VSC-3, the number of concurrent file operations is bound by spinning disk physics: small file (write) operations can easily saturate capacity. Hence, please mind that ''$HOME'' is a shared resource over all projects on a given NFS server. In case your project requires persistence over a large number of small files please contact VSC administration in advance.    
  
-The [[http://www.beegfs.com|BeeGFS]] parallel file system (former FhGFS) on VSC-3 provides a large (initially approx. 0.5 PB) scratch space. The environment variable $GLOBAL expands to: 
  
-<code> 
-$ echo $GLOBAL 
-/global/lv70999/username 
-</code> 
  
-The directory is writeable as user and readable by the group members. It is advisable to make use of these directories in particular for jobs with heavy I/O operations. In addition it will reduce the load on the fileserver holding the $HOME directories. 
- 
-The BeeGFS (former Fraunhofer parallel file system) is shared by all users and by all nodes. 
-Single jobs producing heavy load (>>1000 requests per second) have been observed to reduce responsiveness for all jobs and all users. 
- 
-Lifetime of data is limited, see table below. 
- 
- 
-=== Per-node Scratch Directories $SCRATCH === 
- 
-Local scratch directories on each node are provided as a link to the BeeGFS parallel file system and can thus be viewed also via the login nodes as '''/fhgfs/nodelocal/rXXnXX/''' 
-The parallel file system (and thus the performance) is identical between $SCRATCH and $GLOBAL. 
-The variable ''$SCRATCH'' expands to: 
-<code> 
-$ echo $SCRATCH 
-/scratch 
-</code> 
-These directories are purged after job execution. 
  
 ==== Usage Local Scratch RAM Disk $TMPDIR ==== ==== Usage Local Scratch RAM Disk $TMPDIR ====
  
-'$TMPDIR' provides a small ephermal-volatile RAM disk of 50% node RAM, e.g. 32GB for a 64GB node. It suits very fast local access that is restricted to single nodes, especially for many small files. The RAM disk does not explicitly have to be requested in jobs and grows with file contents - subtracting its usage from available memory. The variable'$TMPDIR' expands to /tmp. Please do not hardcode /tmp directly. Directories in '$TMPDIR' are purged after job execution.+''$TMPDIR'' provides a small ephermal-volatile RAM disk of 50% node RAM, e.g. 32GB for a 64GB node. It suits very fast local access that is restricted to single nodes, especially for many small files. The RAM disk does not explicitly have to be requested in jobs and grows with file contents - subtracting its usage from available memory. The variable ''$TMPDIR'' expands to /tmp. Please do not hardcode /tmp directly. Directories in ''$TMPDIR'' are purged after job execution.
  
 <code> <code>
Line 56: Line 35:
 </code> </code>
  
-\\ 
-==== Comparison of scratch directories ==== 
- 
-|                      || $GLOBAL                        || $SCRATCH               || $TMPDIR      || 
-| Recommended file size || large                         || large                  || small        || 
-| Lifetime             || files older than 90 days deleted if $GLOBAL space is running low     || job                    || job                     || 
-| Size                 || x00 TB (for all users)         || x00 TB (for all users) || a few GB (within memory) ||  
-| Scaling              || does not fit very large number of small file IO || does not fit very large number of small file IO  || very good (local) || 
-| Visibility           || global                         || node (see above)      || node                    ||  
-| Recommended usage     || large files, available temporarily after job life || large files    || many small files (>1000, or many seek-operations within a file ||  
 \\ \\
  
Line 73: Line 42:
  
 Storage extensions can be requested through [[https://service.vsc.tuwien.ac.at/|Vergabeassistent]] at Extensions - Storage. Storage extensions can be requested through [[https://service.vsc.tuwien.ac.at/|Vergabeassistent]] at Extensions - Storage.
- 
-Since 2017-06-02 quotas are enforced for the $GLOBAL filesystem. Additional information can be found [[vsc3_global_quotas|here]] 
  
 ==== Fair Use ==== ==== Fair Use ====
Line 92: Line 59:
 VSC-3 NFS and BeeGFS (former FhGFS) servers utilize RAID-6 that can sustain up to 2 disks failing concurrently. The data path is otherwise not redundant. Data loss may also occur due to failure modes including, but not limited to natural disaster, cooling failure, disk controller failure and filesystem software faults. VSC-3 NFS and BeeGFS (former FhGFS) servers utilize RAID-6 that can sustain up to 2 disks failing concurrently. The data path is otherwise not redundant. Data loss may also occur due to failure modes including, but not limited to natural disaster, cooling failure, disk controller failure and filesystem software faults.
  
-User data on VSC-[[doku:backup|is not backuped]].+User data on VSC-[[doku:backup|is not backuped]].
  
  • doku/vsc3_storage.1496403412.txt.gz
  • Last modified: 2017/06/02 11:36
  • by sreinwal