Data Grid and Gridflow Management Systems Arun swaran

Data Grid and Gridflow Management Systems Arun swaran Jagatheesan San Diego Supercomputer Center (SDSC) University of California at San Diego University of Florida Grid Physics Network (Gri. Phy. N) Diego Supercomputer Center, University of California at San Diego San

“On Demand” Calibration of Content • Been here before? a) b) c) d) e) f) Data / Storage in Grid - Concepts Data / Storage in Grid - Implementation Web Services in Grids Grid Workflow All the above Who would be elected in US? 2 San Diego Supercomputer Center University of Florida University of California at San Diego

Talk Outline • Introduction • Grid Computing, Data Grids • Data Grid Infrastructures • Data Grid Management System (DGMS) • Basic Concepts • Implementation Example (SRB) • Grid Middleware as Service • Gridflows • Related Topics and Research Issues • Demo / Hands on session 3 San Diego Supercomputer Center University of Florida University of California at San Diego

Acknowledgement: SDSC SRB Team Ø Arun Jagatheesan Ø George Kremenek Ø Sheau-Yen Chen Ø Arcot Rajasekar Ø Reagan Moore Ø Michael Wan Ø Roman Olschanowsky Ø Bing Zhu Ø Charlie Cowart Not In Picture: Ø Wayne Schroeder Ø Tim Warnock (BIRN) Ø Lucas Gilbert Ø Marcio Faerman (SCEC) Ø Antoine De Torcy Students: Jonathan Weinberg Yufang Hu Daniel Moore Grace Lin Allen Ding Yi Li Emeritus: Vicky Rowley (BIRN) Qiao Xin Ethan Chen Reena Mathew Erik Vandekieft Xi (Cynthia) Sheng 4 San Diego Supercomputer Center University of Florida University of California at San Diego

Distributed Computing © Images courtesy of Computer History Museum 5 San Diego Supercomputer Center University of Florida University of California at San Diego

The “Grid” Vision 6 San Diego Supercomputer Center University of Florida University of California at San Diego

Using a Data Grid – in Abstract As r fo k d a at D el d Data Grid e er iv • User asks for data/storage from the data grid • The data/storage is found and returned • Where & how details are managed by data grid • But access controls are specified by owner 8 San Diego Supercomputer Center University of Florida University of California at San Diego

Data Grids – Hype Factor/Reality? • Forrester Research • “Data and infrastructure are top of mind for grid at more than 50 percent of firms … The vision of data grids will become part of a greater vision of storage virtualization and information life cycle management” – May 2004 • CIO Magazine • “While most people think of computational grids, enterprises are looking into data grids” – May 2004 • Why talk about Busine$$ in an IEEE conference? • Necessity drives business; business drives standards and technology evolution; …; grid is not just technology, but also standards 9 San Diego Supercomputer Center University of Florida University of California at San Diego

Talk Outline • Introduction • Grid Computing, Data Grids • Data Grid Infrastructures • Data Grid Management System (DGMS) • Basic Concepts • Implementation Example (SRB) • Grid Middleware as Service • Gridflows • Related Topics and Research Issues • Demo / Hands on session 10 San Diego Supercomputer Center University of Florida University of California at San Diego

DGMS Technology usage • • • NSF Southern California Earthquake Center digital library Worldwide Universities Network data grid NASA Information Power Grid NASA Goddard Data Management System data grid DOE Ba. Bar High Energy Physics data grid NSF National Virtual Observatory data grid NSF ROADnet real-time sensor collection data grid NIH Biomedical Informatics Research Network data grid NARA research prototype persistent archive NSF National Science Digital Library persistent archive NHPRC Persistent Archive Test bed 11 San Diego Supercomputer Center University of Florida University of California at San Diego

Southern California Earthquake Center 12 San Diego Supercomputer Center University of Florida University of California at San Diego

Southern California Earthquake Center • Build community digital library • Manage simulation and observational data • Anelastic wave propagation output • 10 TBs, 1. 5 million files • Provide web-based interface • Support standard services on digital library • Manage data distributed across multiple sites • USC, SDSC, UCSB, SDSU, SIO • Provide standard metadata • Community based descriptive metadata • Administrative metadata • Application specific metadata 13 San Diego Supercomputer Center University of Florida University of California at San Diego

SCEC Data Management Technologies • Portals • Knowledge interface to the library, presenting a coherent view of the services • Knowledge Management Systems • Organize relationships between SCEC concepts and semantic labels • Process management systems • Data processing pipelines to create derived data products • Web services • Uniform capabilities provided across SCEC collections • Data grid • Management of collections of distributed data • Computational grid • Access to distributed compute resources • Persistent archive • Management of technology evolution 14 San Diego Supercomputer Center University of Florida University of California at San Diego

15 San Diego Supercomputer Center University of Florida University of California at San Diego

NASA Data Grids • NASA Information Power Grid • NASA Ames, NASA Goddard • Distributed data collection using the SRB • ESIP federation • Led by Joseph Ja. Ja (U Md) • Federation of ESIP data resources using the SRB • NASA Goddard Data Management System • Storage repository virtualization (Unix file system, Unitree archive, DMF archive) using the SRB • NASA EOS Petabyte store • Storage repository virtualization for EMC persistent store using the Nirvana version of SRB 16 San Diego Supercomputer Center University of Florida University of California at San Diego

Tera. Grid: 13. 6 TF, 6. 8 TB memory, 900 TB network disk, 10 PB archive Caltech 0. 5 TF. 4 TB Memory 86 TB disk 32 32 24 Calren 256 p HP X-Class 24 128 p HP V 2500 24 1 TF. 25 TB Memory 25 TB disk 4 574 p IA-32 Chiba City 32 32 92 p IA-32 8 8 NTON ANL Extreme Blk Diamond 5 HPSS 32 128 p Origin 32 HR Display & VR Facilities 5 HPSS OC-48 OC-12 ATM OC-12 Chicago & LA DTF Core Switch/Routers Cisco 65 xx Catalyst Switch (256 Gb/s Crossbar) Juniper M 160 OC-48 OC-12 Gb. E v. BNS Abilene Calren ESnet SDSC OC-12 4. 1 TF 2 TB Memory 500 TB SAN OC-12 OC-3 NCSA 6+2 TF 4 TB Memory 400 TB disk ESnet HSCC MREN/Abilene Starlight OC-12 OC-3 4 8 HPSS Uni. Tree 8 9 PB 4 Blue Horizon 2 Sun Server 1176 p IBM SP 1. 7 TFLOPs 1024 p IA-32 320 p IA-64 14 16 4 v. BNS Abilene MREN Myrinet 2 x Sun E 10 K 17 San Diego Supercomputer Center University of Florida University of California at San Diego 15 xxp Origin

NIH BIRN SRB Data Grid • Biomedical Informatics Research Network • Access and analyze biomedical image data • Data resources distributed throughout the country • Medical schools and research centers across the US • Stable high performance grid based environment • Coordinate data sharing • Federate collections • Support data mining and analysis 18 San Diego Supercomputer Center University of Florida University of California at San Diego

BIRN: Inter-organizational Data 19 San Diego Supercomputer Center University of Florida University of California at San Diego

SRB Collections at SDSC 20 San Diego Supercomputer Center University of Florida University of California at San Diego

Commonality in all these projects • Distributed data management • Data Grids, Digital Libraries, Persistent Archives, • Workflow/dataflow Pipelines, Knowledge Generation • Data/storage provisioning – multiple domains • Common logical namespace for data and storage • Data publication • Browsing and discovery of data in collections • Data Preservation • Management of technology evolution 21 San Diego Supercomputer Center University of Florida University of California at San Diego

Talk Outline • Introduction • Grid Computing, Data Grids • Data Grid Infrastructures • Data Grid Management System (DGMS) • Basic Concepts • Implementation Example (SRB) • Grid Middleware as Service • Gridflows • Related Topics and Research Issues • Demo / Hands on session 22 San Diego Supercomputer Center University of Florida University of California at San Diego

Physical Layer (Real World) • Distributed digital entities • Heterogeneous and distributed storage resources • Autonomous Organizations • Distributed Users, distributed authentication • Heterogeneous authorization schemes • Users; sub-organizations; organizations/enterprises; virtual organizations 24 San Diego Supercomputer Center University of Florida University of California at San Diego

Data Grid Transparencies/Virtualizations (bits, data, information, . . ) Inter. Semantic data Organization (with behavior) my. Active. Neuro. Collection patient. Records. Collection Virtual Data Transparency image. cgi image. wsdl image. sql Data Replica Transparency image_0. jpg…image_100. jpg organizational Information Storage Management Data Identifier Transparency E: srb. Vaultimage. jpg /users/srb. Vault/image. jpg Select … from srb. mdas. td where. . . Storage Location Transparency Storage Resource Transparency 25 San Diego Supercomputer Center University of Florida University of California at San Diego

Data Grid Transparencies • Find data without knowing the identifier • Descriptive attributes • Access data/storage without knowing the location • Logical name space • Access data without knowing the type of storage • Storage repository abstraction • Provide transformations for any data collection • Data behavior abstraction 26 San Diego Supercomputer Center University of Florida University of California at San Diego

Data Grid Abstractions • Storage repository virtualization • Standard operations supported on storage systems • Data virtualization • Logical name space for files - Global persistent identifier • Information repository virtualization • Standard operations to manage collections in databases • Access virtualization • Standard interface to support alternate APIs • Latency management mechanisms • Aggregation, parallel I/O, replication, caching • Security interoperability • GSSAPI, inter-realm authentication, collection-based authorization 27 San Diego Supercomputer Center University of Florida University of California at San Diego

Data Organization • Physical Organization of the data • Distributed Data • Heterogeneous resources • Multiple formats (structured and unstructured) • Logical Organization • Impose logical structure for data sets • Collections of semantically related data sets • Users create their own views (collections) of the data grid 28 San Diego Supercomputer Center University of Florida University of California at San Diego

Data Identifier Transparency Four Types of Data Identifiers: • Unique name • OID or handle • Descriptive name • Descriptive attributes – meta data • Semantic access to data • Collective name • Logical name space of a collection of data sets • Location independent • Physical name • Physical location of resource and physical path of data 29 San Diego Supercomputer Center University of Florida University of California at San Diego

Mappings on Resource Name Space • Define logical resource name • List of physical resources • Replication • Write to logical resource completes when all physical resources have a copy • Load balancing • Write to a logical resource completes when copy exist on next physical resource in the list • Fault tolerance • Write to a logical resource completes when copies exist on “k” of “n” physical resources 30 San Diego Supercomputer Center University of Florida University of California at San Diego

Data Replica Transparency • Replication • • Improve access time Improve reliability Provide disaster backup and preservation Physically or Semantically equivalent replicas • Replica consistency • Synchronization across replicas on writes • Updates might use “m of n” or any other policy • Distributed locking across multiple sites • Versions of files • Time-annotated snapshots of data 31 San Diego Supercomputer Center University of Florida University of California at San Diego

Latency Management -Bulk Operations • Bulk register • Create a logical name for a file • Bulk load • Create a copy of the file on a data grid storage repository • Bulk unload • Provide containers to hold small files and pointers to each file location • Bulk delete • Mark as deleted in metadata catalog • After specified interval, delete file • Bulk metadata load • Requests for bulk operations for access control setting, … 32 San Diego Supercomputer Center University of Florida University of California at San Diego

Talk Outline • Introduction • Grid Computing, Data Grids • Data Grid Infrastructures • Data Grid Management System (DGMS) • Basic Concepts • Implementation Example (SRB) • Grid Middleware as Service • Gridflows • Related Topics and Research Issues • Demo / Hands on session 33 San Diego Supercomputer Center University of Florida University of California at San Diego

Storage Resource Broker • Distributed data management technology • Developed at San Diego Supercomputer Center (Univ. of California, San Diego) • 1996 - DARPA Massive Data Analysis • 1998 - DARPA/USPTO Distributed Object Computation Test bed • 2000 to present - NSF, NASA, NARA, DOE, DOD, NIH, NLM, NHPRC • Applications • • Data grids - data sharing Digital libraries - data publication Persistent archives - data preservation Used in national and international projects in support of Astronomy, Bio-Informatics, Biology, Earth Systems Science, Ecology, Education, Geology, Government records, High Energy Physics, Seismology 34 San Diego Supercomputer Center University of Florida University of California at San Diego

Data Grid Federation • Data grids provide the ability to name, organize, and manage data on distributed storage resources • Federation provides a way to name, organize, and manage data on multiple data grids. 35 San Diego Supercomputer Center University of Florida University of California at San Diego

SRB Zones • Each SRB zone uses a metadata catalog (MCAT) to manage the context associated with digital content • Context includes: • • Administrative, descriptive, authenticity attributes Users Resources Applications 36 San Diego Supercomputer Center University of Florida University of California at San Diego

SRB Peer-to-Peer Federation • Mechanisms to impose consistency and access constraints on: • Resources • Controls on which zones may use a resource • User names (user-name / domain / SRB-zone) • Users may be registered into another domain, but retain their home zone, similar to Shibboleth • Data files • Controls on who specifies replication of data • MCAT metadata • Controls on who manages updates to metadata 37 San Diego Supercomputer Center University of Florida University of California at San Diego

Peer-to-Peer Federation 1. Occasional Interchange - for specified users 2. Replicated Catalogs - entire state information replication 3. Resource Interaction - data replication 4. Replicated Data Zones - no user interactions between zones 5. Master-Slave Zones - slaves replicate data from master zone 6. Snow-Flake Zones - hierarchy of data replication zones 7. User / Data Replica Zones - user access from remote to home zone 8. Nomadic Zones “SRB in a Box” - synchronize local zone to parent 9. Free-floating “my. Zone” - synchronize without a parent zone 10. Archival “Back. Up Zone” - synchronize to an archive SRB Version 3. 0. 1 released December 19, 2003 38 San Diego Supercomputer Center University of Florida University of California at San Diego

Talk Outline • Introduction • Grid Computing, Data Grids • Data Grid Infrastructures • Data Grid Management System (DGMS) • Basic Concepts • Implementation Example (SRB) • Grid Middleware as Service • Gridflows • Related Topics and Research Issues • Demo / Hands on session 39 San Diego Supercomputer Center University of Florida University of California at San Diego

Why Data Grids Need Services? • Standardization • Grid vision can not be achieved without standards • Dynamic discovery and late-binding of resources on demand • Resources as Services • Each resource could be treated as an endpoint • Flexibility • XML internally helps in interoperable resource descriptions 40 San Diego Supercomputer Center University of Florida University of California at San Diego

Grid Middleware as Services • Standard protocols / interfaces • Heterogeneous implementation infrastructure • Standard Port. Type for a Data Grid Service Provider • Standard Schemes • Flexible open standards for operations on Data Grids • Adding an organization to a data grid programmatically • Adding resources on demand • Standard Data models for Resources (Schemas) • Description of Resources in a Grid • Resources become Services 41 San Diego Supercomputer Center University of Florida University of California at San Diego

Challenges in using Services for Data Grids • High Performance Requirement • Can verbose XML between clients and grid middleware? • Scalability • Will XML based services scale or become bottleneck? • Wait for SOA Standardization • (e. g) WSDL 2. 0 - necessary wait though • Community Resistance • “I will use the regular API – at least for now” • Tools for development 42 San Diego Supercomputer Center University of Florida University of California at San Diego

Standardization of Data Grid Services • Global Grid Forum (GGF) • Data Area • Grid File System, DAIS, GSM, … • Open Grid Services Architecture (OGSA) • WS-Resource Framework • Slowly evolving as a community effort 43 San Diego Supercomputer Center University of Florida University of California at San Diego

GFS Service Provider /home/arun. sdsc/exp 1/text 1. txt /home/arun. sdsc/exp 1/text 2. txt /home/arun. sdsc/exp 1/text 3. txt data + storage (100) Research Lab data + storage (10) Storage-R-Us Resource Providers data + storage (50) GRP GRP /…/text 1. txt Logical Namespace (Need not be same as physical view of resources ) Finance Department data + storage (40) GRP GRP /…//text 2. txt GRP /txt 3. txt 44 San Diego Supercomputer Center University of Florida University of California at San Diego GRP

Talk Outline • Introduction • Grid Computing, Data Grids • Data Grid Infrastructures • Data Grid Management System (DGMS) • Basic Concepts • Implementation Example (SRB) • Grid Middleware as Service • Gridflows • Related Topics and Research Issues • Demo / Hands on session 45 San Diego Supercomputer Center University of Florida University of California at San Diego

Work in progress Gf. MS is ‘Hard Hat Area’ (Research) 46 San Diego Supercomputer Center University of Florida University of California at San Diego

Gridflow in SCEC (data information pipeline) Metadata derivation Ingest Data Ingest Metadata Pipeline could be triggered by input at data source or by a data request from user Determine analysis pipeline Initiate automated analysis Use the optimal set of resources based on the task – on demand Organize result data into distributed data grid collections All gridflow activities stored for data flow provenance 47 San Diego Supercomputer Center University of Florida University of California at San Diego

Gridflows • Grid Workflow (Gridflow) is the automation of a execution pipeline in which data or tasks are processed through multiple autonomous grid resources according to a set of procedural rules • Gridflows are executed on resources that are dynamically obtained through confluence of one or more autonomous administrative domains (peers) 48 San Diego Supercomputer Center University of Florida University of California at San Diego

Data Discovery New data Digital entities updates relationships among data in collections Meta-data Services invoked to analyze new relationships Services DGMS applications get notified of state updates State 49 San Diego Supercomputer Center University of Florida University of California at San Diego

What they want? We know the business (scientific) process Cyber. Infrastructure is all we care (why bother about atoms or DNA) 50 San Diego Supercomputer Center University of Florida University of California at San Diego

What they want? Use DGL to describe your process logic with abstract references to datagrid infrastructure dependencies 51 San Diego Supercomputer Center University of Florida University of California at San Diego

Need for Gridflows • Data-intensive and/or compute-intensive processes • Long run processes or pipelines on the Grid • (e. g) If job A completes execute jobs x, y, z; else execute job B. • Self-organization/management of data • Semi-automation of data, storage distribution, curation processes • (e. g) After each data insert into a collection, update the meta-data information about the collection or replicate the collection • Knowledge Generation • Offline data analysis and knowledge generation pipelines 52 San Diego Supercomputer Center University of Florida • (e. g) What inferencesof California at San Diego from the new University can be assumed

Gridflow Description Requirements • Import and export • Import or export Gridflows (embedded gridflows) • Support and extend existing standards like XQuery, BPEL, SOAP etc. , • Rules • Dynamic rules to control the execution of gridflow • Query • Runtime Query on status of gridflow • Granular Metadata • Metadata associated with the steps in a gridflow execution that can be queried • Gridflow Patterns • Scientific Computing - more looping structures • Interest in execution of each iteration and the changes in interested attributes 53 San Diego Supercomputer Center • http: //tmitwww. tm. tue. nl/research/patterns/standards. htm University of Florida University of California at San Diego

Data Grid Language • Assembly Language for Grid Computing? (ok, its hype) • Describes Gridflow • Both structure-based and state-based gridflow patterns • Described ECA based rules • Inbuilt support to define data grid datatypes like collections, … • Query Gridflow • Query on the execution of any gridflow (any granular detail) • XQuery is used to query on the status of gridflow and its attributes 54 San Diego Supercomputer Center University of Florida University of California at San Diego • Manage Gridflow

• Structure and state based Gridflow patterns Simple Sequential • Execute steps in a gridflow in a sequence one after another • Simple Parallel • Start all the steps in a gridflow at the same time • For Loop Iteration • Execute steps changing some iterator value until a given state is achieved • While Block (Milestone) • Execute steps while some mile stone can be achieved • IF-Else Block • Branch based on the evaluation of a state condition • Switch-choice(s) • Split to execute any of the possible cases based on the context • For-each 55 San Diego Supercomputer Center University of Florida University of California at San Diego

SDSC Matrix Project • R&D effort that is used in academic projects • Gridflow Protocols • Gridflow Language Descriptions • Version 3. 2 released, implements DGL • Community based • Both Industry and Academia can benefit by participation • Involves University of Florida, UCSD, … (Are you In? ) 56 San Diego Supercomputer Center University of Florida University of California at San Diego

Gridflow Process I End User using DGBuilder Gridflow Description Data Grid Language 57 San Diego Supercomputer Center University of Florida University of California at San Diego

Gridflow Process II Abstract Gridflow using Data Grid Language Planner Concrete Gridflow 58 San Diego Supercomputer Center University of Florida University of California at San Diego

Gridflow Process III Gridflow Processor Concrete Gridflow P 2 P Network 59 San Diego Supercomputer Center University of Florida University of California at San Diego

SDSC Matrix Project: Open source gridflow effort • The growth of the SDSC Matrix Project is made possible by developers and grid-prophets like you (Thank you) • talk 2 Matrix@sdsc. edu 60 San Diego Supercomputer Center University of Florida University of California at San Diego

Talk Outline • Introduction • Grid Computing, Data Grids • Data Grid Infrastructures • Data Grid Management System (DGMS) • Basic Concepts • Implementation Example (SRB) • Grid Middleware as Service • Gridflows • Related Topics and Research Issues • Demo / Hands on session 61 San Diego Supercomputer Center University of Florida University of California at San Diego

DGMS Philosophy • Collective view of • Inter-organizational data • Operations on datagrid space • Local autonomy and global state consistency • Collaborative datagrid communities • Multiple administrative domains or “Grid Zones” • Self-describing and self-manipulating data • Horizontal and vertical behavior • Loose coupling between data and behavior (dynamically) • Relationships between a digital entity and its Physical locations, Logical names, Meta-data, Access control, Behavior, “Grid Zones”. 62 San Diego Supercomputer Center University of Florida University of California at San Diego

DGMS Research Issues • Self-organization of datagrid communities • Using knowledge relationships across the datagrids • Inter-datagrid operations based on semantics of data in the communities (different ontologies) • High speed data transfer • Terabyte to transfer • Protocols, routers needed • Latency Management • Data source speed >> data sink speed • Datagrid Constraints • Data placement and scheduling • How many replicas, where to place them… 63 San Diego Supercomputer Center University of Florida University of California at San Diego

Active Datagrid Collections Resources Data Sets 121. Event Behavior Thit. xml 121. Event get. Events() National Lab Hits. sql add. Event() SDSC University of Gators 64 San Diego Supercomputer Center University of Florida University of California at San Diego

Active Datagrid Collections 121. Event Thit. xml Heterogeneous, distributed physical data 121. Event get. Events() National Lab Dynamic or virtual data Hits. sql add. Event() SDSC University of Gators 65 San Diego Supercomputer Center University of Florida University of California at San Diego

Active Datagrid Collections Logical Collection gives location and naming transparency my. HEP-Collection Meta-data 121. Event Thit. xml National Lab 121. Event SDSC Hits. sql University of Gators 66 San Diego Supercomputer Center University of Florida University of California at San Diego

Active Datagrid Collections Now add behavior or services to this logical collection Collection state and services my. HEP-Collection Meta-data Horizontal Services 121. Event Thit. xml 121. Event get. Events() National Lab Hits. sql add. Event() SDSC University of Gators 67 San Diego Supercomputer Center University of Florida University of California at San Diego

Active Datagrid Collections ADC Logical view of data & operations ADC specific Operations + Model View Controllers Collection state and services my. HEP-Collection Meta-data Horizontal Services 121. Event Thit. xml 121. Event get. Events() National Lab Hits. sql add. Event() SDSC University of Gators 68 San Diego Supercomputer Center University of Florida University of California at San Diego

Active Datagrid Collections Physical and virtual data present in the datagrid Digital entities Meta-data Services State Standardized schema with domain specific schema extensions Horizontal datagrid services and vertical domain specific services (port. Type) or pipelines (DGL) Events, collective state, mappings to domain services to be invoked 69 San Diego Supercomputer Center University of Florida University of California at San Diego

Talk Outline • Introduction • Grid Computing, Data Grids • Data Grid Infrastructures • Data Grid Management System (DGMS) • Basic Concepts • Implementation Example (SRB) • Grid Middleware as Service • Gridflows • Related Topics and Research Issues • Demo / Hands on session 70 San Diego Supercomputer Center University of Florida University of California at San Diego

Related Technologies/Links – Data Grids • • A complete history of the Grid SDSC Storage Resource Broker Globus Data Grid The Legion Project To go a link, use them in Google’s “I’m Feeling Lucky” option (or look into the notes in this presentation) 71 San Diego Supercomputer Center University of Florida University of California at San Diego

Talk Outline • Introduction • Grid Computing, Data Grids • Data Grid Infrastructures • Data Grid Management System (DGMS) • Basic Concepts • Implementation Example (SRB) • Grid Middleware as Service • Gridflows • Related Topics and Research Issues • Demo / Hands on session 72 San Diego Supercomputer Center University of Florida University of California at San Diego

Summary • Data Grids – next generation data management • • Lot of possibilities and use cases Transparencies for distributed storage and data Data Grid Management System Need for standard Services • Gridflows • Peer-2 -peer Grid Workflows • Research Issues 73 San Diego Supercomputer Center University of Florida University of California at San Diego