Difference between revisions of "Swestore-dCache"

From SNIC Documentation
Jump to: navigation, search
(Download and upload data)
(The JavaScript WebDAV client is dead/unusable now, remove.)
 
(179 intermediate revisions by 10 users not shown)
Line 1: Line 1:
 
[[Category:Storage]]
 
[[Category:Storage]]
[[Category:SweStore]]
+
[[Category:Swestore]]
SNIC is building a storage infrastructure to complement the computational resources.
+
[[Category:Swestore]]
 +
[[Category:Swestore user guide]]
  
Many forms of automated measurements can produce large amounts of data. In scientific areas such as high energy physics (the Large Hadron Collider at CERN), climate modeling, bioinformatics, bioimaging etc., the demands for storage are increasing dramatically. To serve these and other user communities, SNIC has appointed a working group to design a storage strategy, taking into account the needs on many levels and creating a unified storage infrastructure, which is now being implemented.
+
Swestore-dCache is distributed across the SNIC centres [http://www.c3se.chalmers.se/ C3SE], [http://www.hpc2n.umu.se/ HPC2N], [http://www.lunarc.lu.se/ Lunarc] and [http://www.nsc.liu.se/ NSC]. Data is stored in two copies with each copy at a different SNIC centre. This enables the system to cope with a multitude of issues ranging from a simple crash of a storage element to losing an entire site while still providing access to the stored data.  
  
Swestore is in collaboration with [http://www.ecds.se/ ECDS], [http://snd.gu.se/ SND], Bioimage Sweden, [http://www.bils.se/ BILS], [http://www.uppnex.uu.se/ UPPNEX],[http://wlcg.web.cern.ch/ WLCG], [http://www.nrm.se/ NaturHistoriska RiksMuseet].
+
One of the major advantages to the distributed nature of Swestore is the excellent aggregated transfer rates possible. This is achieved by bypassing a central node and having transfers going directly to/from the storage elements if the selected transfer protocol allows it. Swestore can achieve aggregated transfer rates in excess of 100 Gigabit per second, but in practice this is limited by connectivity to the end user, each university or a limited number of files (typically max 1 Gbit/s per file/connection).
  
= National storage =
+
To protect against silent data corruption the dCache storage system checksums all stored data and periodically verifies the data using this checksum.
The aim of the nationally accessible storage is to build a robust, flexible and expandable system that can
 
be used in most cases where access to large scale storage is needed. To the user it should appear as a single large system,
 
while it is desirable that some parts of the system are distributed across all SNIC centra to benefit from the advantages
 
of, among other things, locality and cache effects. The system is intended as a versatile long-term storage system.
 
  
==Supported access protocol==
+
The dCache system does NOT yet provide protection against user errors like inadvertent file deletions.
; Today SweStore support this protocols
 
: srm://, gsiftp://, http:// (ro), https:// (ro), webdav (rw).
 
; Coming to support this protocols
 
: NFS4.1, iRODS
 
  
== Getting access ==
+
== Basic information ==
; Apply for storage
 
: Please follow instructions [[Apply for storage on SweStore|here]]
 
; Get a client certificate.
 
: Follow the instructions [[Grid_certificates#Requesting_a_certificate|here]] to get your client certificate. For Terena certificates, please make sure you also [[Requesting_a_grid_certificate_using_the_Terena_eScience_Portal#Exporting Terena certificate for use with Grid tools|export the certificate for use with grid tools]]. For Nordugrid certificates, please make sure to also [[Requesting_a_grid_certificate_from_the_Nordugrid_CA#Installing_the_certificate_in_your_browser|install your client certificate in your browser]].
 
; Request membership in the SweGrid VO.
 
: Follow the instructions [[Grid_certificates#Requesting_membership_in_the_SweGrid_VO|here]] to get added to the SweGrid virtual organisation.
 
  
 +
For basic information on how to apply for storage or become a member of an existing project and managing your storage username/password, please check the main [[Swestore]] page.
 +
 +
== Acquire and manage certificate (for users accessing with certificate authentication) ==
 +
 +
Certificate authentication is needed in some special cases, and more convenient for use in automation (ie. Robot certificates).
 +
 +
Some communities require certificate authentication and such users might already have a certificate, that certificate is likely good to use with Swestore.
 +
 +
If you need to use Certificate authentication, follow the instruction on the page [[Certificate Setup for Swestore]].
 +
 +
= Using Swestore =
 
== Download and upload data ==
 
== Download and upload data ==
; Interactive browsing and manipulation of single files
+
=== From the command line ===
: SweStore is accessible from your web browser, here https://webdav.swegrid.se/. To browse private data you must first install your certificate in your browser (see above). Your data is available at <code><nowiki>https://webdav.swegrid.se/snic/YOUR_PROJECT_NAME</nowiki></code>.
+
 
; Upload and delete data interactively or with automation
+
There are several command line tools capable of using the protocols provided by Swestore. For usage on SNIC clusters we recommend using the ARC tools which should be installed on all SNIC resources.
There are several tools that are capable of using the protocols provided by SweStore national storage.
+
 
For interactive usage on SNIC clusters and resources we recommend using the ARC tools.
 
 
As an integration point for building scripts and automated systems we suggest using the curl program and library.
 
As an integration point for building scripts and automated systems we suggest using the curl program and library.
: Use the ARC client. Please see the instructions for [[Accessing SweStore national storage with the ARC client]].
 
: Use cURL. Please see the instructions for [[Accessing SweStore national storage with cURL]].
 
: Use lftp. Please see the instructions for [[Accessing SweStore national storage with lftp]].
 
: Use globus-url-copy. Please see the instructions for [[Accessing SweStore national storage with globus-url-copy]].
 
  
== More information ==
+
:; Interactive tools
* [[SweStore introduction]]
+
:: Best suited for interactive use with various support for directory listings, wildcards etc.
* [http://status.swestore.se/munin/monitor/monitor/ Per Project Monitoring of Swestore usage]
+
::;Currently no suitable tools to recommend
* [[Accessing SweStore national storage with the ARC client]]
+
:; Command-line tools
<!-- * [[Mounting SweStore national storage via WebDAV|Mounting SweStore national storage via WebDAV (Not recomendated at the moment)]] -->
+
:: These tools are also suitable for scripting or automation.
If you have any issues using SweStore please do not hesitate to contact [mailto:swestore-support@snic.vr.se swestore-support].
+
::; rclone
 +
::: Supports multiple protocols.
 +
::: Please see the instructions for [[Accessing Swestore with rclone]].
 +
::; cURL
 +
::: Please see the instructions for [[Accessing Swestore with cURL]]. We suggest using this as integration point for building scripts and automated systems.
 +
::; davix
 +
::: Swestore-specific documentation not written yet. See upstream documentation at https://davix.web.cern.ch/davix/docs/master/
 +
::: Use access URL: davs://webdav.swestore.se:1443/
 +
::; ARC
 +
::: Certificate authentication only. Supports multiple protocols.
 +
::: Please see the instructions for [[Accessing Swestore with the ARC client]].
 +
::; gfal2
 +
::: Certificate authentication only. Supports multiple protocols.
 +
::: Swestore-specific documentation not written yet. See upstream documentation at https://dmc.web.cern.ch/projects/gfal-2/documentation
 +
::: Use access URL: gsiftp://gsiftp.swestore.se/
 +
::; https://duck.sh/
 +
::: Command-line client for Windows/macOS/Linux (not documented yet)
 +
 
 +
=== Using a GUI client ===
 +
 
 +
Graphical User Interface (GUI) clients are known to work on some operating systems.
 +
 
 +
:; Cyberduck
 +
:: Please see the instructions for [[Accessing Swestore using Cyberduck]] '''Recommended method on macOS'''
 +
:; WinSCP
 +
:: Please see the instructions for [[Accessing Swestore using WinSCP]] '''Recommended method on Windows'''
 +
 
 +
=== From a web browser ===
 +
 
 +
Swestore is accessible in your web browser as a simple directory index interface at https://webdav.swestore.se/.
 +
 
 +
Web browser access supports both username/password and certificate authentication.
 +
 
 +
Projects are organized under the <code>/snic</code> directory as <code><nowiki>https://webdav.swestore.se/snic/DIRECTORY_NAME/</nowiki></code>.
 +
 
 +
=== Legacy documentation ===
  
== Tools and scripts ==
+
There are other clients/methods available, but not recommended for new deployments. See [[:Category:Swestore user guide]] for a complete list of documentation.
  
There exists a number of tools and utilities developed externally that can be useful. Here are some links:
+
== Enabled access protocols ==
 +
A design criteria for Swestore is to provide the storage over a number of standardized and public protocols. There is no vendor specific client needed for access.  
  
* [https://github.com/samuell/arc_tools ARC_Tools] - Convenience scripts for the arc client (Only a recursive rmdir so far).
+
:; GridFTP
* [http://sourceforge.net/projects/arc-gui-clients ARC Graphical Clients] - Contains the ARC Storage Explorer (SweStore supported development).
+
:: Also called gsiftp. Well supported within Swestore.
* Transfer script, [http://snicdocs.nsc.liu.se/wiki/SweStore/swstrans_arc swetrans_arc], provided by Adam Peplinski / Philipp Schlatter
+
:: Features: Transfer checksums. Direct transfer from/to storage pools for high speed transfers.
* [http://www.nordugrid.org/documents/SWIG-wrapped-ARC-Python-API.pdf Documentation of the ARC Python API (PDF)]
+
:: Access URL: gsiftp://gsiftp.swestore.se/
 +
:; HTTP/WebDAV
 +
:: Contender for being the recommended protocol for Swestore.
 +
:: Features: Support in web browsers. Supports direct transfer of files from/to storage pools improving performance (varying support in clients).
 +
::; Authenticated access
 +
::: This is the normal read-write access.
 +
::: Access URL: https://webdav.swestore.se/ (non-redirected)
 +
::: Access URL: https://webdav.swestore.se:1443/ (redirected)
 +
::: Access URL: https://webdav.swestore.se:2443/ (redirected reads)
 +
::; Unauthenticated read-only access
 +
::: For special cases only.
 +
::: Access URL: http://webdav.swestore.se/ (unauthenticated, read-only, non-redirected)
 +
::: Access URL: http://webdav.swestore.se:1080/ (unauthenticated, read-only, redirected)
 +
::: Access URL: http://webdav.swestore.se:2080/ (unauthenticated, read-only, redirected reads)
 +
:; NFSv4.1
 +
:: Used by some communities, not recommended for general use
 +
:; SRM - Storage Resource Manager
 +
:: Used by some communities, not recommended for general use.
 +
:: Access URL: srm://srm.swegrid.se/
 +
:; Xrootd
 +
:: Used by some communities, not recommended for general use
 +
:; DCAP
 +
:: Used by some communities, not recommended for general use
  
= Center storage =
+
== Usage monitoring ==
Centre storage, as defined by the SNIC storage group, is a storage solution that lives independently of the computational resources and can be accessed from all such resources at a centre. Key features include the ability to access the same filesystem the same way on all computational resources at a centre, and a unified structure and nomenclature for all centra. Unlike cluster storage which is tightly associated with a single cluster, and thus has a limited life-time, centre storage does not require the users to migrate their own data when clusters are decommissioned, not even when the storage hardware itself is being replaced.
+
* On the project page in SUPR: https://supr.snic.se
 +
* On the monitoring server for Swestore: http://status.swestore.se/munin/monitor/monitor/  (only accessible from .se-domains.
  
== Unified environment ==
+
= Support =
To make the usage more transparent for SNIC users, a set of environment variables are available on all SNIC resources:
 
  
* <code>SNIC_BACKUP</code> – the user's primary directory at the centre<br>(the part of the centre storage that is backed up)
+
If you have any issues using Swestore, please do not hesitate to contact [mailto:support@swestore.se support@swestore.se].
* <code>SNIC_NOBACKUP</code> – recommended directory for project storage without backup<br>(also on the centre storage)
 
* <code>SNIC_TMP</code> – recommended directory for best performance during a job<br>(local disk on nodes if applicable)
 

Latest revision as of 09:42, 20 October 2021


Swestore-dCache is distributed across the SNIC centres C3SE, HPC2N, Lunarc and NSC. Data is stored in two copies with each copy at a different SNIC centre. This enables the system to cope with a multitude of issues ranging from a simple crash of a storage element to losing an entire site while still providing access to the stored data.

One of the major advantages to the distributed nature of Swestore is the excellent aggregated transfer rates possible. This is achieved by bypassing a central node and having transfers going directly to/from the storage elements if the selected transfer protocol allows it. Swestore can achieve aggregated transfer rates in excess of 100 Gigabit per second, but in practice this is limited by connectivity to the end user, each university or a limited number of files (typically max 1 Gbit/s per file/connection).

To protect against silent data corruption the dCache storage system checksums all stored data and periodically verifies the data using this checksum.

The dCache system does NOT yet provide protection against user errors like inadvertent file deletions.

Basic information

For basic information on how to apply for storage or become a member of an existing project and managing your storage username/password, please check the main Swestore page.

Acquire and manage certificate (for users accessing with certificate authentication)

Certificate authentication is needed in some special cases, and more convenient for use in automation (ie. Robot certificates).

Some communities require certificate authentication and such users might already have a certificate, that certificate is likely good to use with Swestore.

If you need to use Certificate authentication, follow the instruction on the page Certificate Setup for Swestore.

Using Swestore

Download and upload data

From the command line

There are several command line tools capable of using the protocols provided by Swestore. For usage on SNIC clusters we recommend using the ARC tools which should be installed on all SNIC resources.

As an integration point for building scripts and automated systems we suggest using the curl program and library.

Interactive tools
Best suited for interactive use with various support for directory listings, wildcards etc.
Currently no suitable tools to recommend
Command-line tools
These tools are also suitable for scripting or automation.
rclone
Supports multiple protocols.
Please see the instructions for Accessing Swestore with rclone.
cURL
Please see the instructions for Accessing Swestore with cURL. We suggest using this as integration point for building scripts and automated systems.
davix
Swestore-specific documentation not written yet. See upstream documentation at https://davix.web.cern.ch/davix/docs/master/
Use access URL: davs://webdav.swestore.se:1443/
ARC
Certificate authentication only. Supports multiple protocols.
Please see the instructions for Accessing Swestore with the ARC client.
gfal2
Certificate authentication only. Supports multiple protocols.
Swestore-specific documentation not written yet. See upstream documentation at https://dmc.web.cern.ch/projects/gfal-2/documentation
Use access URL: gsiftp://gsiftp.swestore.se/
https://duck.sh/
Command-line client for Windows/macOS/Linux (not documented yet)

Using a GUI client

Graphical User Interface (GUI) clients are known to work on some operating systems.

Cyberduck
Please see the instructions for Accessing Swestore using Cyberduck Recommended method on macOS
WinSCP
Please see the instructions for Accessing Swestore using WinSCP Recommended method on Windows

From a web browser

Swestore is accessible in your web browser as a simple directory index interface at https://webdav.swestore.se/.

Web browser access supports both username/password and certificate authentication.

Projects are organized under the /snic directory as https://webdav.swestore.se/snic/DIRECTORY_NAME/.

Legacy documentation

There are other clients/methods available, but not recommended for new deployments. See Category:Swestore user guide for a complete list of documentation.

Enabled access protocols

A design criteria for Swestore is to provide the storage over a number of standardized and public protocols. There is no vendor specific client needed for access.

GridFTP
Also called gsiftp. Well supported within Swestore.
Features: Transfer checksums. Direct transfer from/to storage pools for high speed transfers.
Access URL: gsiftp://gsiftp.swestore.se/
HTTP/WebDAV
Contender for being the recommended protocol for Swestore.
Features: Support in web browsers. Supports direct transfer of files from/to storage pools improving performance (varying support in clients).
Authenticated access
This is the normal read-write access.
Access URL: https://webdav.swestore.se/ (non-redirected)
Access URL: https://webdav.swestore.se:1443/ (redirected)
Access URL: https://webdav.swestore.se:2443/ (redirected reads)
Unauthenticated read-only access
For special cases only.
Access URL: http://webdav.swestore.se/ (unauthenticated, read-only, non-redirected)
Access URL: http://webdav.swestore.se:1080/ (unauthenticated, read-only, redirected)
Access URL: http://webdav.swestore.se:2080/ (unauthenticated, read-only, redirected reads)
NFSv4.1
Used by some communities, not recommended for general use
SRM - Storage Resource Manager
Used by some communities, not recommended for general use.
Access URL: srm://srm.swegrid.se/
Xrootd
Used by some communities, not recommended for general use
DCAP
Used by some communities, not recommended for general use

Usage monitoring

Support

If you have any issues using Swestore, please do not hesitate to contact support@swestore.se.