1
The ESGF.org is a community driven, open-development effort to produce robust, open-sourced infrastructures for data distribution. These systems are conceived to support data- intensive science by guaranteeing performant and transparent access to Peta/Exa-scale scientific data in a geographically distributed federation. The primary application is the ESGF Node, an amalgam of software components that work in concert to perform the higher level tasks. The Node is a collection of components developed by the members of the ESGF.org open source effort as well as external tools and libraries that provide support for the ESGF Node feature set. The ESGF Node will be used to support the CMIP5/IPCC AR5 data infrastructure having 1 PB of replicated data and 10 PB in total across the federation. The ESGF Node allows an institution to check, publish, replicate and administer data as well as to secure access to its resources. The modularity of the ESGF Node is central to its conception and it already provides means for file serving (HTTP, GridFTP, OpenDAP), browsing (Thredds), visualization, access metrics, as well as many others. Register... user@server:~$ globus-url-copy gsiftp://server/somefile ~/myfile Source: gsiftp://server:2811/ Dest: file:///home/ somefile -> myfile 1310720 bytes 20.10 MB/sec avg 24.00 MB/sec inst user@server:~$ wget_script --2011-03-21 13:58:44-- https://server/somefile Resolving server... 128.128.128.128 Connecting to server|128.128.128.128|:443... connected. HTTP request sent, awaiting response... 200 OK Length: 76643715 (73M) [application/gzip] Saving to: `somefile' 6% [==> ] 4,855,572 1.52M/s eta 45s ...download... ...find... ...or even visualize & compute! Earth System Grid Federation .org (ESGF.org) ESGF Node Estanislao Gonzalez (1), Gavin M. Bell (2), Luca Cinquini (3), Philip Kershaw (4), Stephan Kindermann (5), Michael Lautenschlager (5), Stephen Pascoe (4), and Dean Williams (2) (1) Max-Planck-Institut für Meteorologie, Data Management, Hamburg, Germany ([email protected]) (2) Program for Climate Model Diagnosis and Intercomparison, Lawrence Livermore National Labs (3) Jet PropulsionLaboratory, National Aeronautics and Space Administration (4) NCAS/British Atmospheric Data Centre, RutherfordAppleton Laboratory, STFC (5) Deutsches Klimarechenzentrum (DKRZ) CMIP5 / IPCC AR5 A data Infrastructure for data-intensive science Security Search Compute Data Manager Estanislao Gonzalez Max-Planck-Institut für Meteorologie (MPI-M) Phone: +49 (40) 46 00 94-126 E-Mail: [email protected] Contact http://pcmdi3.llnl.gov/esgcet/ http://cmip-gw.badc.rl.ac.uk/ http://ipcc-ar5.dkrz.de/ ... The ESGF.org web presence is http://esgf.org For information regarding hosted data instead of the infrastructure holding it please contact: [email protected]

ESGF Node A data Infrastructure for data-intensive science

  • Upload
    others

  • View
    2

  • Download
    0

Embed Size (px)

Citation preview

Page 1: ESGF Node A data Infrastructure for data-intensive science

The ESGF.org is a community driven, open-development effort

to produce robust, open-sourced infrastructures for data

distribution. These systems are conceived to support data-

intensive science by guaranteeing performant and transparent

access to Peta/Exa-scale scientific data in a geographically

distributed federation.

The primary application is the ESGF Node, an amalgam of

software components that work in concert to perform the

higher level tasks. The Node is a collection of components

developed by the members of the ESGF.org open source effort

as well as external tools and libraries that provide support for

the ESGF Node feature set.

The ESGF Node will be used to support the CMIP5/IPCC AR5

data infrastructure having 1 PB of replicated data and 10 PB in

total across the federation. The ESGF Node allows an

institution to check, publish, replicate and administer data as

well as to secure access to its resources. The modularity of the

ESGF Node is central to its conception and it already provides

means for file serving (HTTP, GridFTP, OpenDAP), browsing

(Thredds), visualization, access metrics, as well as many others.

Register...

user@server:~$ globus-url-copy gsiftp://server/somefile ~/myfile

Source: gsiftp://server:2811/

Dest: file:///home/

somefile -> myfile

1310720 bytes 20.10 MB/sec avg 24.00 MB/sec inst

user@server:~$ wget_script

--2011-03-21 13:58:44-- https://server/somefile

Resolving server... 128.128.128.128

Connecting to server|128.128.128.128|:443... connected.

HTTP request sent, awaiting response... 200 OK

Length: 76643715 (73M) [application/gzip]

Saving to: `somefile'

6% [==> ] 4,855,572 1.52M/s eta 45s

...download......find... ...or even visualize & compute!

Earth System Grid Federation .org(ESGF.org)

ESGF Node

Estanislao Gonzalez (1), Gavin M. Bell (2), Luca Cinquini (3), Philip Kershaw (4), Stephan Kindermann (5), Michael Lautenschlager (5), Stephen Pascoe (4), and Dean Williams (2)(1) Max-Planck-Institut für Meteorologie, Data Management, Hamburg, Germany ([email protected])

(2) Program for Climate Model Diagnosis and Intercomparison, Lawrence Livermore National Labs (3) Jet PropulsionLaboratory, National Aeronautics and Space Administration

(4) NCAS/British Atmospheric Data Centre, RutherfordAppleton Laboratory, STFC (5) Deutsches Klimarechenzentrum (DKRZ)

CMIP5 / IPCC AR5

A data Infrastructure for data-intensive science

Security

Search

Compute

Data

Manager

Estanislao Gonzalez

Max-Planck-Institut für Meteorologie (MPI-M)

Phone: +49 (40) 46 00 94-126

E-Mail: [email protected]

Contact

http://pcmdi3.llnl.gov/esgcet/

http://cm

ip-gw.badc.rl.ac.uk/

http://ipcc-ar5.dkrz.de/

...

The ESGF.org web presence is http://esgf.org

For information regarding hosted data instead of

the infrastructure holding it please contact:

[email protected]