MDS-2.1 and Futures Karl Czajkowski Information Sciences Institute University of Southern California.

Slides:



Advertisements
Similar presentations
Giggle: A Framework for Constructing Scalable Replica Location Services Ann Chervenak, Ewa Deelman, Ian Foster, Leanne Guy, Wolfgang Hoschekk, Adriana.
Advertisements

The Replica Location Service In wide area computing systems, it is often desirable to create copies (replicas) of data objects. Replication can be used.
MDS-2.1 and Futures Karl Czajkowski Information Sciences Institute University of Southern California.
Data Management Expert Panel - WP2. WP2 Overview.
The Anatomy of the Grid: An Integrated View of Grid Architecture Carl Kesselman USC/Information Sciences Institute Ian Foster, Steve Tuecke Argonne National.
FP7-INFRA Enabling Grids for E-sciencE EGEE Induction Grid training for users, Institute of Physics Belgrade, Serbia Sep. 19, 2008.
Indications in green = Live content Indications in white = Edit in master Indications in blue = Locked elements Indications in black = Optional elements.
Connect. Communicate. Collaborate Click to edit Master title style MODULE 1: perfSONAR TECHNICAL OVERVIEW.
Condor-G: A Computation Management Agent for Multi-Institutional Grids James Frey, Todd Tannenbaum, Miron Livny, Ian Foster, Steven Tuecke Reporter: Fu-Jiun.
A Computation Management Agent for Multi-Institutional Grids
Grid Information Service (Meta-Directory Service 2) Globus Toolkit™ Developer Tutorial The Globus Project™ Argonne National Laboratory USC Information.
USING THE GLOBUS TOOLKIT This summary by: Asad Samar / CALTECH/CMS Ben Segal / CERN-IT FULL INFO AT:
 Introduction Originally developed by Open Software Foundation (OSF), which is now called The Open Group ( Provides a set of tools and.
Evaluation of MDS 2.1 at Saclay D. Calvet, Z. Georgette, M. Huet, J-P. Le Fèvre, I. Mandjavidze, P. Micout, B. Thooris DAPNIA/SEI, CEA Saclay Gif-sur-Yvette.
Massimo Cafaro GridLab Review GridLab WP10 Information Services Massimo Cafaro CACT/ISUFI University of Lecce, Italy.
Grid Computing, B. Wilkinson, 20046c.1 Globus III - Information Services.
Systems Architecture, Fourth Edition1 Internet and Distributed Application Services Chapter 13.
INFN experience with Globus GIS A. Cavalli - F. Semeria INFN Grid Information Services workshop CERN, March 2001.
Web-based Portal for Discovery, Retrieval and Visualization of Earth Science Datasets in Grid Environment Zhenping (Jane) Liu.
Grid Monitoring By Zoran Obradovic CSE-510 October 2007.
Distributed Computing COEN 317 DC2: Naming, part 1.
Grid Toolkits Globus, Condor, BOINC, Xgrid Young Suk Moon.
Grid Information Systems. Two grid information problems Two problems  Monitoring  Discovery We can use similar techniques for both.
Kuali Rice at Indiana University Rice Setup Options July 29-30, 2008 Eric Westfall.
The EU DataGrid – Information and Monitoring Services The European DataGrid Project Team
GRID Centralized management of the Globus grid-mapfile Carlo Rocca INFN, Catania.
Overview of the NorduGrid Information System Balázs Kónya 3 rd NorduGrid Workshop 23 May, 2002, Helsinki.
Grid Discovery and Monitoring Systems Laura Pearlman USC/Information Sciences Institute With materials from Ben Clifford and others from the Globus Project.
GT Components. Globus Toolkit A “toolkit” of services and packages for creating the basic grid computing infrastructure Higher level tools added to this.
1 School of Computer, National University of Defense Technology A Profile on the Grid Data Engine (GridDaEn) Xiao Nong
A. Cavalli - F. Semeria INFN Experience With Globus GIS 1 A. Cavalli - F. Semeria INFN First INFN Grid Workshop Catania, 9-11 April 2001 INFN Experience.
Distributed Computing COEN 317 DC2: Naming, part 1.
© Oxford University Press 2011 DISTRIBUTED COMPUTING Sunita Mahajan Sunita Mahajan, Principal, Institute of Computer Science, MET League of Colleges, Mumbai.
EGEE-II INFSO-RI Enabling Grids for E-sciencE EGEE and gLite are registered trademarks Information System on gLite middleware Vincent.
A monitoring tool for a GRID operation center Sergio Andreozzi (INFN CNAF), Sergio Fantinel (INFN Padova), David Rebatto (INFN Milano), Gennaro Tortone.
1 Schema Registries Steven Hughes, Lou Reich, Dan Crichton NASA 21 October 2015.
1 4/23/2007 Introduction to Grid computing Sunil Avutu Graduate Student Dept.of Computer Science.
October 27, 2015 Atlas Monitoring Infrastructure in Grid Environment Richard Baker Dantong Yu Brookhaven National Lab.
Introduction to Grid Computing Ed Seidel Max Planck Institute for Gravitational Physics
The Replica Location Service The Globus Project™ And The DataGrid Project Copyright (c) 2002 University of Chicago and The University of Southern California.
NA-MIC National Alliance for Medical Image Computing UCSD: Engineering Core 2 Portal and Grid Infrastructure.
What is SAM-Grid? Job Handling Data Handling Monitoring and Information.
Overview of Privilege Project at Fermilab (compilation of multiple talks and documents written by various authors) Tanya Levshina.
Globus – Part II Sathish Vadhiyar. Globus Information Service.
E-infrastructure shared between Europe and Latin America FP6−2004−Infrastructures−6-SSA gLite Information System Pedro Rausch IF.
Introduction to Grids By: Fetahi Z. Wuhib [CSD2004-Team19]
PPDG February 2002 Iosif Legrand Monitoring systems requirements, Prototype tools and integration with other services Iosif Legrand California Institute.
FP6−2004−Infrastructures−6-SSA E-infrastructure shared between Europe and Latin America gLite Information System Claudio Cherubino.
GraDS MacroGrid Carl Kesselman USC/Information Sciences Institute.
Super Computing 2000 DOE SCIENCE ON THE GRID Storage Resource Management For the Earth Science Grid Scientific Data Management Research Group NERSC, LBNL.
The EU DataGrid – Information and Monitoring Services The European DataGrid Project Team
EGI-Engage Data Services and Solutions Part 1: Data in the Grid Vincenzo Spinoso EGI.eu/INFN Data Services.
Grid Information Services The Globus Project Argonne National Laboratory USC Information Sciences Instutute
GIIS Implementation and Requirements F. Semeria INFN European Datagrid Conference Amsterdam, 7 March 2001.
The Globus Toolkit The Globus project was started by Ian Foster and Carl Kesselman from Argonne National Labs and USC respectively. The Globus toolkit.
FESR Trinacria Grid Virtual Laboratory gLite Information System Muoio Annamaria INFN - Catania gLite 3.0 Tutorial Trigrid Catania,
A System for Monitoring and Management of Computational Grids Warren Smith Computer Sciences Corporation NASA Ames Research Center.
E-science grid facility for Europe and Latin America Updates on Information System Annamaria Muoio - INFN Tutorials for trainers 01/07/2008.
Vincenzo Spinoso EGI.eu/INFN
File System Implementation
Globus —— Toolkits for Grid Computing
Network Load Balancing
Introduction to Data Management in EGI
DNS.
The Globus Toolkit™: Information Services
Introduction to Name and Directory Services
JINI ICS 243F- Distributed Systems Middleware, Spring 2001
Introduction to Active Directory Directory Services
Lecture 4: File-System Interface
Presentation transcript:

MDS-2.1 and Futures Karl Czajkowski Information Sciences Institute University of Southern California

Talk Outline l Introduction –Problem, history, etc. l MDS-2 Architecture –Protocols –Features, services l MDS-2.1 Software –Technology map l Information Model –Additional background

Resource Discovery/Monitoring l Distributed users and resources l Variable resource status l Variable grouping and connectivity R R R R R R ? ? R R R R RR R RR ? ? R R R R R dispersed users VO-AVO-B partioned network

Basic Grid Acquisition Phases l Resource Discovery –“What resources are relevant?” –Bootstraps planner state l Resource Status Query –“How do resources compare (now)?” –Refines planner knowledge l Resource Control –“Did I acquire the resources?” –Not an information service task!

MDS History l MDS-1 (classic) –Centralized database –Globus and earlier –Did not scale l MDS-2 –Distributed services –MDS 2.0 in Globus –New MDS 2.1 development in alpha release

Base Features l Virtual Organizations (Vos) –Group together resources –Support community-specific “discovery” –Specialized “views” l Scalability –Many resources –Many Vos –Graceful degradation of service

Virtual Organizations l Collaborating individuals and institutions –Shared goals –Enable sharing of resources –Non-locality of participants l Dynamic in nature –VOs come and go –Resources joing and leave Vos –Resource change status and fail l Community-wide goals

Scalability l Large numbers –Many resources –Many users l Independence –Resources shouldn’t affect one another –Vos shouldn’t affect one another l Graceful degradation of service –“As much function as possible” –Tolerate partitions, prune failures

New MDS-2.1 Features l Security –GSI mutual-authentication –Fine-grained access control by GSI name l Performance –Better query speeds –Less stale information l Extensibility l Convenience

Service Hierarchy l Resource Description via Info. Protocol (GRIP) –Co-located with resource on network l Aggregate Directories (via GRIP or other) –Can be made hierarchical l Dynamic Registration via Reg. Protocol (GRRP) discovery (GRIP?) lookup (GRIP) registration (GRRP) VO-specific AggDirs standard ResDesc services RRRR AA?

Distributed Services l Service scales with Grid growth l Loose consistency model tolerates failures l Interoperability by protocols D D D D

Soft-state Registration l Periodic notification –Service/resource is available –Expected-frequency metadata l Automatic extension –Add new resources to directories –Invite resource to join new directory l Self-cleaning –Reduce occurrence of “dead” references

MDS-2 Implementation l Grid Resource Information Service (GRIS) –Provides resource description –Modular content gateway l Grid Index Information Service (GIIS) –Provides aggregate directory –Hierarchical groups of resources l Lightweight Dir. Access Protocol (LDAP) –Standard with many client implementations –Used for GRIP (and GRRP currently)

MDS-2.1 Development Activities l Incorporating external advances –New OpenLDAP 2.0.x code-base –Cyrus-SASL/GSI security integration –Leveraging new Globus packaging model l Improving internal components –Better query servicing –New configuration/policy support –Invitation (reverse registration)

MDS-2.1 External Software Stack l OpenLDAP 2.0.x (.11) –Implements LDAPv3 protocol –Client and server components l Cyrus-SASL –Generic security –We provide loadable SASL/GSI plugin l Globus GSI (repackaged) –Provides GSS-API interface to PKI –Loadable module works with SASL plugin

MDS 2.1 Security l PKI authentication l Static authorization –Class, attribute, object name rules l “Self” authorization –Semi-dynamic rule –Requires “owner” attribute on objects l Dynamic authorization –Directory-based group lists (or future CAS) –Per-object access rule attributes

MDS-2.1 Internal Software l Wrappers/tools –Simplify typical idioms l Modular GRIS providers –Probe/query resource status –Generates LDIF-format data l LDAP server “backend” modules –GRIS provider dispatch/caching –GIIS implementation(s)

MDS-2.1alpha GRIS Providers globus-software reports Globus packages grid-info-host reports host OS info grid-info-host-interfaces reports NICs grid-info-host-load reports host load grid-info-host-filesystem reports disks globus-gram-reporter reports jobs

GRIS Dispatch Tests l Concurrent dispatch for each provider: 1.Could search intersect provider? No, then stop. 2.Is provider cache stale? Yes, then refill. 3.Apply search filter to cache data. l Combine all providers’ results

MDS-2.1 GRIS Configuration dn: sw=Globus, hn=${GLOBUS_HOSTNAME}, ${GRID_INFO_ ORGANIZATION_DN} objectclass: GlobusTop objectclass: GlobusActiveObject objectclass: GlobusActiveSearch type: exec path: /opt/globus-mds/bin base: globus-version args: -ldif cachetime: timelimit: 10 sizelimit: 1 …

GRIS Configuration cont’d dn: hn=${GLOBUS_HOSTNAME},{GRID_INFO_ORGANIZ ATION_DN} objectclass: GlobusTop objectclass: GlobusActiveObject objectclass: GlobusActiveSearch type: exec path: /opt/globus-mds/libexec base: globus-gram-reporter args: -f /opt/globus-mds/etc/globus-gram-rep orter.conf -onetime cachetime: 30 timelimit: 10 sizelimit: 20 …

Hierarchical GIIS l Maintain set of remote services –Track incoming live registrations –GRIS or GIIS registrants l Cached proxy results (now), or –Same cache logic as GRIS –Refill cache with “chaining” queries l LDAPv3 referral results (planned) –Do not maintain any local info cache –Redirect clients to active registrants

Extensible GIIS Framework l Modular registration actions –Re-use registration protocol decoding –Specialize directory update –e.g. prefetch indexable data l Modular query actions –Re-use query protocol decoding –Specialize query handler algorithm –e.g. utilize precomputed indices

MDS-LDAP Data Model l Info named within service l Info tagged with content type name(s) l Values associated with typed attributes values types name objectclass: computer system: mips irix dn: hn=hostX objectclass: service dispatchtype: immediate dn: queue=default, hn=hostX url: gram://hostname/default objectclass: queue dn: perf=load5, hn=hostX objectclass: perf objectclass: loadaverage period: 10 load5: 3.2 dn: store=scratch, hn=hostX objectclass: storage objectclass: raidstore free: MB raidmode: stripe

MDS-LDAP Query Model l Search scoping –Search rooted in namespace –Search depth of “root,” “root’s children,” or “root’s subtree” l Search filter –Value or type comparison –Logical combinations of filters l Namespace represents concept space

Namespace Management l Info is named uniquely within a service l Append “source name” to disambiguate locally, or use URLs to refer to remote info host host: hn=R3 host: hn=R2 host: hn=R1 host: hn=R2 host: hn=R1 host: hn=R2, O=O2 host: hn=R1, O=O2 host: hn=R3, O=O1 O1O2R1 R2R1R3R2R1 host: hn=R2, O=O1 AggDir ResDesc host: hn=R1, O=O1

More Information l Questions? l HPDC-10 Paper (to appear August 2001) –“Grid Information Services for Distributed Resource Sharing” l MDS-2.1 Alpha Website – –Early access to development code