Riccardo Zappi INFN-CNAF SRM Breakout session. February 28, 2012 Ingredients 1. Basic ingredients (Fabric & Conn. level) 2. (Grid) Middleware ingredients.

Slides:



Advertisements
Similar presentations
© 2007 Open Grid Forum Data Management Challenge - The View from OGF OGF22 – February 28, 2008 Cambridge, MA, USA Erwin Laure David E. Martin Data Area.
Advertisements

Jens G Jensen Atlas Petabyte store Supporting Multiple Interfaces to Mass Storage Providing Tape and Mass Storage to Diverse Scientific Communities.
Data Management Expert Panel. RLS Globus-EDG Replica Location Service u Joint Design in the form of the Giggle architecture u Reference Implementation.
Workflows over Grid-based Web services General framework and a practical case in structural biology gLite 3.0 Data Management David García Aristegui Grid.
EGEE-II INFSO-RI Enabling Grids for E-sciencE The gLite middleware distribution OSG Consortium Meeting Seattle,
EGEE is a project funded by the European Union under contract IST Using SRM: DPM and dCache G.Donvito,V.Spinoso INFN Bari
The EPIKH Project (Exchange Programme to advance e-Infrastructure Know-How) gLite Grid Services Abderrahman El Kharrim
EGEE-II INFSO-RI Enabling Grids for E-sciencE gLite Data Management System Yaodong Cheng CC-IHEP, Chinese Academy.
INFSO-RI Enabling Grids for E-sciencE gLite Data Management Services - Overview Mike Mineter National e-Science Centre, Edinburgh.
Data management in grid. Comparative analysis of storage systems in WLCG.
LHCb input to DM and SM TEGs. Remarks to DM and SM TEGS Introduction m We have already provided some input during our dedicated session of the TEG m Here.
Data Management The GSM-WG Perspective. Background SRM is the Storage Resource Manager A Control protocol for Mass Storage Systems Standard protocol:
EGEE-II INFSO-RI Enabling Grids for E-sciencE EGEE middleware Data Management in gLite.
D C a c h e Michael Ernst Patrick Fuhrmann Tigran Mkrtchyan d C a c h e M. Ernst, P. Fuhrmann, T. Mkrtchyan Chep 2003 Chep2003 UCSD, California.
Author - Title- Date - n° 1 Partner Logo EU DataGrid, Work Package 5 The Storage Element.
EGEE-II INFSO-RI Enabling Grids for E-sciencE EGEE middleware: gLite Data Management EGEE Tutorial 23rd APAN Meeting, Manila Jan.
Enabling Grids for E-sciencE Introduction Data Management Jan Just Keijser Nikhef Grid Tutorial, November 2008.
From Digital Objects to Content across eInfrastructures Content and Storage Management in gCube Pasquale Pagano CNR –ISTI on behalf of Heiko Schuldt Dept.
E-science grid facility for Europe and Latin America Data Management Services E2GRIS1 Rafael Silva – UFCG (Brazil) Universidade Federal.
INFSO-RI Enabling Grids for E-sciencE The gLite File Transfer Service: Middleware Lessons Learned form Service Challenges Paolo.
Managing Data DIRAC Project. Outline  Data management components  Storage Elements  File Catalogs  DIRAC conventions for user data  Data operation.
Scientific Storage at FNAL Gerard Bernabeu Altayo Dmitry Litvintsev Gene Oleynik 14/10/2015.
INFSO-RI Enabling Grids for E-sciencE Introduction Data Management Ron Trompert SARA Grid Tutorial, September 2007.
Padova, 5 October StoRM Service view Riccardo Zappi INFN-CNAF Bologna.
David Adams ATLAS ATLAS distributed data management David Adams BNL February 22, 2005 Database working group ATLAS software workshop.
EGEE-II INFSO-RI Enabling Grids for E-sciencE EGEE and gLite are registered trademarks gLite Data Management Components Presenter.
EGI-Engage Data Services and Solutions Part 1: Data in the Grid Vincenzo Spinoso EGI.eu/INFN Data Services.
EGEE-II INFSO-RI Enabling Grids for E-sciencE Data management in EGEE.
Andrea Manzi CERN EGI Conference on Challenges and Solutions for Big Data Processing on cloud 24/09/2014 Storage Management Overview 1 24/09/2014.
INFSO-RI Enabling Grids for E-sciencE University of Coimbra gLite 1.4 Data Management System Salvatore Scifo, Riccardo Bruno Test.
INFSO-RI Enabling Grids for E-sciencE University of Coimbra Data Management System gLite – LCG – FiReMan Salvatore Scifo INFN Catania.
EGEE-II INFSO-RI Enabling Grids for E-sciencE Architecture of LHC File Catalog Valeria Ardizzone INFN Catania – EGEE-II NA3/NA4.
The EPIKH Project (Exchange Programme to advance e-Infrastructure Know-How) Algiers, EUMED/Epikh Application Porting Tutorial, 2010/07/04.
Enabling Grids for E-sciencE EGEE-II INFSO-RI Status of SRB/SRM interface development Fu-Ming Tsai Academia Sinica Grid Computing.
Introduction to Storage Element Hsin-Wei Wu Academia Sinica Grid Computing Center, Taiwan.
Martedi 8 novembre 2005 Consorzio COMETA “Progetto PI2S2” FESR Data Management System Annamaria Muoio -- INFN Catania PI2S2 First Tutorial -- Messina,
EGI-InSPIRE RI EGI-InSPIRE EGI-InSPIRE RI EGI solution for high throughput data analysis Peter Solagna EGI.eu Operations.
Enabling Grids for E-sciencE EGEE-II INFSO-RI The Development of SRM interface for SRB Fu-Ming Tsai Academia Sinica Grid Computing.
EGEE Data Management Services
a brief summary for users
CASTOR: possible evolution into the LHC era
GFAL Grid File Access Library
GFAL Grid File Access Library
GFAL: Grid File Access Library
gLite Basic APIs Christos Filippidis
Ricardo Rocha ( on behalf of the DPM team )
ATLAS Use and Experience of FTS
StoRM: a SRM solution for disk based storage systems
Vincenzo Spinoso EGI.eu/INFN
Status of the SRM 2.2 MoU extension
gLite Data Management Services
gLite Data management system overview
GGF OGSA-WG, Data Use Cases Peter Kunszt Middleware Activity, Data Management Cluster EGEE is a project funded by the European.
Storage Interfaces and Access: Introduction
StoRM Architecture and Daemons
Gfal/lcg-util -> Gfal2/gfal2-util
Introduction to Data Management in EGI
Taming the protocol zoo
Introduction to reading and writing files in Grid
Grid Services Ouafa Bentaleb CERIST, Algeria
Hands-On Session: Data Management
GFAL 2.0 Devresse Adrien CERN lcgutil team
The INFN Tier-1 Storage Implementation
Data Management cluster summary
Data Management Ouafa Bentaleb CERIST, Algeria
Data services in gLite “s” gLite and LCG.
Architecture of the gLite Data Management System
gLite Data and Metadata Management
INFNGRID Workshop – Bari, Italy, October 2004
StoRM disk management middleware
Presentation transcript:

Riccardo Zappi INFN-CNAF SRM Breakout session

February 28, 2012 Ingredients 1. Basic ingredients (Fabric & Conn. level) 2. (Grid) Middleware ingredients  Site level middleware (Resource level)  VO level middleware (Collective level)  Security aspects (Orthogonal level) 3. Experiment Application ingredients (Application level) 2

Fabric and Connectivity 3 January 24-25, 2011, TEG DM & SM, NIKHEF, Amsterdam

February 28, 2012 Basic Ingredients: Hardware  Hardware Solid State drive (SSD), Hard Disk, Tape Storage System, … SATA, SCSI, Fiber Channel, … NAS, DAS, SAN and TAN Hierarchical storage solutions: 2 tiers: SAN and TAN 3 or more tiers. 4

February 28, 2012  File Systems  Lustre, GPFS: commercial products. Great, but the support is not for free.  Hadoop FS: used by big (very big) company; To adapt the usage (Map/Reduce) to exploit the HDFS features …  CEPH FS: not yet in production quality. not used by no-one “Big Company”; sustainability issue? needed a direct involvement Man power issue? Basic Ingredients: File Systems 5

February 28, 2012  File Data Access protocol  Native POSIX  pNFS (NFSv4.1 protocol)  FUSE access  xrootd  RFIO  dCap  … Basic Ingredients: connectivity 6

Resource 7 January 24-25, 2011, TEG DM & SM, NIKHEF, Amsterdam

February 28, 2012 Site level: Storage Resource  Data are ultimately stored in a storage system:  from memory stick to a multi-petabytes tape silo  Different storage resources offer different levels of Quality of Service (QoS).  The Storage Resource is the combination of the storage hardware and the controlling software (eg. file system) 8

February 28, 2012 Site level: Storage Element  Storage Element is the service which allows a user or an application to store data for future retrieval.  Offer two kind of services:  Data Access/ Transfer services  Storage Resource Manager (SRM) services 9

February 28, 2012 Site level: Data Access/Transfer Services  Data Access/Transfer services http(s) standard, widely used and supported WebDav standard, widely used and supported GridFTP OGF standard, extensively used in Grid communities Xrootd non standard, used exclusively by (some) HEP communities RFIO non standard, used exclusively by (some) HEP communities 10

February 28, 2012 Uniform interface to heterogeneous Storage Resources 11

February 28, 2012 Site level: SRM service  Storage Resource Manager (SRM) service  Hide the complexity of storage resource: such as hierarchical storage system (hsm)  Manage storage space and files reserve space, volatile files, move files between tiers  Hide local file namespace SURL (Storage URL): logical URL used to reference the file TURL (Transfer URL): physical URL used to access the file 12

February 28, 2012 Storage Element in a site 13

February 28, 2012 Storage Element solutions (1/2)  Storage Element  Two kinds of SE: All-inclusive (DPM and dCache) Assemble-as-you-need (StoRM and BestMan)  Data Transfer Protocol  GridFTP (obviosly)  http(s), BitTorrent  other (GridTorrent ?) 14

February 28, 2012 Storage Element solutions (2/2)  Data Access Protocol  Standard compliant protocols pNFS (NFSv4.1), WebDav, file  Specific protocol (non standard) xrootd, rfio, dcap, …  SRM service  All-in-one solution: dCache and DPM Doesn’t support file protocol, but they’ll support pNFS  StoRM Enable native file protocol to exploit of storage system solutions based on Lustre and GPFS.  BestMan Similar to StoRM (in features). OSG distribution 15

Collective Level (VO) 16 StoRM-Tutorial for supporter, IGI, Bologna, Italy

February 28, 2012 “VO level” ingredients  File Catalogue Service  Maintain the mapping between LFNs, GUID, SURLs  Maintain System metadata data regarding the file, not about the content of file  Metadata Service  Maintain metadata about (the content of) files  Data Discovery  Enable the discovery of data (LFNs) using queries on Metadata  Data Movement Service  Create replicas of files  Data Placement Service  Create replicas request to Data Movement Service 17

File Naming and Catalog 18

February 28, 2012 “VO level”: File Catalogue (1/2)  Used to locate Replica locations  Queries are simple (CRUD):  "get replica localities“  “create-new/delete replica”  Existing solution:  LFC 19

February 28, 2012 File Transfer Service (FTS)  gLite File Transfer Service (FTS)  Manage the network and the storage at both ends  Define the concept of a CHANNEL: a link between two SEs  File movement is asynchronous : submit a job held in file transfer queue 20

File Transfer Service (FTS) 21

February 28, 2012 Data management clients gfal/LCG-Utils  Data management access libs.  Shield users from complexity  Interacts with information system, catalogue and SRM-SEs  GFAL  Posix like C API for file access  SRMv2.2 support  lcg_util (command line + C API )  Replication, catalogue interaction etc. 22

February 28, 2012 GFAL: open a file Markus Schulz 23

February 28, 2012 Conclusion StoRM-Tutorial for supporter, IGI, Bologna, Italy 24  Storage Element services interact with many other services and client.  Failures can happen at every level:  Fabric & connectivity: hw failures, File system unresponsive, etc.  Resource: AuthN & AuthZ issues, misconfiguration, firewall, etc.  Connectivity: Information System unreachable, network latency and failures, etc.  End user observing failures don’t know where the failure happen, but she knows only that something was wrong and she is willing to bet that it's a StoRM failure.