the tivoli storage manager in the large hardron patrick...

20
Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005 dCache.ORG dCache.ORG The Tivoli Storage Manager in the Large Hardron Collider Grid world Patrick Fuhrmann for the dCache people

Upload: duongkhuong

Post on 25-Jul-2019

214 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

The Tivoli Storage Manager in the Large Hardron Collider Grid world

Patrick

Fuhrmann for the dCache people

Page 2: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

Tier 0 (CERN)

Tier 2 Centers

IN2P3(Lyon),FR

Karlsruhe,GR

Madrid,SP

Brookhaven,USFermiLab,US RAL,UK

Regional Centers

others ***

Tokio,JP

2.5 – 10 GBit/sec2.5 GBit/sec

LCG Tier Center Mechanism

Aachen

MunichDESY

Italy

Spain

US

Canada

Japan

e.t.c.

Amsterdam, NL

Netherlands

Disk only

Disk and TapeTier 1 Centers

Page 3: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORGLCG Tier Center Mechanism (cont.)

Workload ManagerResource Broker

Compute Element

Storage Element

LCG Site

Compute Element

Storage Element

LCG Site

Compute Element

Storage Element

LCG Site

Compute Element

Storage Element

LCG Site

Compute Element

Storage Element

LCG Site

Compute Element

Storage Element

LCG Site

Job Submission

Job

Data Flow

Page 4: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORGTier 0/1 Compute and Storage Element Setup

Compute Element (CE) Storage Element (SE)

Disk Storage Tape Storage

Tier 1 Center A

Local Access Prot.RFIO, dCap, xrootd

Worker Nodes

Remote Protocols Steering : File Transfer Service (FTS) Control : Storage Resource Manager (SRM)Transfer : GsiFTP / GridFTP

Compute Element (CE) Storage Element (SE)

Disk Storage Tape Storage

Tier 1 Center B

Local Access Prot.RFIO, dCap, xrootd

Worker Nodes

Page 5: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORGStorage Element Requirements

Being a LCG Storage Element, means

support of the Storage Resource Manager Protocol (SRM)

support of the GsiFTP data transfer protocol

support of the GRIS information provider protocol

support of local (posix like) access protocol : rfio, dCap, (xrootd)

Secure Wide Area data transfer protocol

Space allocationTransfer Protocol Negotiation Dataset pinning

Checksum management

Third Party Transfer on behave of dataset owner

Availability Free/Available SpaceAccess Protocol and access details (URIs)

Page 6: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

The dCache Storage Element

Page 7: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORGThe dCache Storage Element

Single 'rooted' file system name space tree

Supports multiple internal and external copies of a single file

Data is distributed among a huge amount of possible cheap disk servers.

Basic Specification

File system names space view available through an nfs2/3 interface

dCache : Managed Disk Storage System

Supports 'posix like' (authenticated) access as well as various FTPdialects and the Storage Resource Manager Protocol.

Page 8: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

Automatic load balancing using cost metric and inter pool transfers.

Distributed Movers AND Access Points (Doors)

Scalability

Pool 2 Pool transfers on pool hot spot detection

Handles bunch requests by fast pool selection unit

dCache : Scalability

Page 9: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

Automatic HSM migration and restore

Pool to pool transfers on configuration of forbidden transfers

Fine grained configuration of pool attraction scheme.(write pools, subnet, directory tree, storage info)

Convenient HSM connectivity for enstore, osm, TSM, Hpss

Configuration

Tertiary Storage Manager connectivity

Fine grained tuning : Space vs. Mover cost preference

HSM dCache interface by script (shell, perl ...)

dCache : Configuration, HSM

Page 10: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

Controls number of copies for each dCache dataset

Makes sure n < copies < m

Adjusts replica count on pool failures

Adjusts replica count on scheduled pool maintenance

Embedded farm node dCache (makes use of local disk space)

Resilient dCache

Resilient dCache

Page 11: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

Local Access Protocol : dCap, Posix like access

SRM version ~ 1 (1.7) plus Space Reservation

limited GRIS functionality, will improve soon

GsiFtp support

dCache : Storage Element

LCG Storage Element

Page 12: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

DESYFnal

dCache Collaboration

Grid KAKarlruhe

LCG Deployment

CERN

Brookhaven

LCG Tier I / II

CDF

US CMS

gridPP Deployment

RAL

LCG Tier I / II UK

San Diego Super Computer Center Virtual Organization Management

INFN BariInformation Provider

Jane ????

dCache development and support Topology

Page 13: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

(gsi,kerberos) dCap Server

Resilient Manager

Ftp Server (gsi, kerberos)

Storage Resource Mgr (SRM)

dCap Client

dCache Core

Cell PackagePnfs TSM Adapter

Gris

HSM Flush Manager

Core Layer

Extended

central services

http / https

AccessLayer

GridLayer

HSM Flush Manager Prestager

dCache module Topology

Page 14: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORGHSM interactions

HSM Interactions

Page 15: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

precious

cached

cached

dCacheTSM

Client

Space needed

File requested

Dataset life cycle

Page 16: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

Data FlowClient -> dCache

dCache -> HSM

Time

Data Transferred

Tape Mount

Deferred HSM flush

Data collected per Storage Class (files space, subdirectory trees)

Collected data flushed following rules (space,time,# of files)

Page 17: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

Scalability by highly distributed Storage Pools and direct client – disk and disk – hsm connection

Client(s)

http(s)

GridFtp

dCapNo transfer funnels or bottlenecks

dCache transfer scalability

Page 18: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORGTSM, the ideal dCache tape backend

The TSM is the ideal dCache tape storage backend

In place at all big lab's (mostly for backup)Mature and well established software

Drivers available for large set of tape libraries

Administration knowledge : wide spread

Already in place (as dCache back end) :

GridKa, KarlruheSara, Amsterdam via SGIs DMF

Canada, TRIUMF Jülich : planned for this year

Page 19: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORGBig Question

Does TSM scale with the upcoming requirements ?

And how ???

Page 20: The Tivoli Storage Manager in the Large Hardron Patrick ...tsm-symposium.oucs.ox.ac.uk/2005/papers/The Tivoli Storage Manager in... · Patrick Fuhrmann TSM Symposium, Oxford Sep 27,

Patrick Fuhrmann TSM Symposium, Oxford Sep 27, 2005

dCache.ORGdCache.ORG

www.dCache.ORG