Core Grid Functions: A Minimal Architecture for Grids William E. Johnston Lawrence Berkeley National Lab and NASA Ames Research Center (www-itg.lbl.gov/~wej)

Slides:



Advertisements
Similar presentations
LEAD Portal: a TeraGrid Gateway and Application Service Architecture Marcus Christie and Suresh Marru Indiana University LEAD Project (
Advertisements

Abstraction Layers Why do we need them? –Protection against change Where in the hourglass do we put them? –Computer Scientist perspective Expose low-level.
GT 4 Security Goals & Plans Sam Meder
High Performance Computing Course Notes Grid Computing.
This product includes material developed by the Globus Project ( Introduction to Grid Services and GT3.
SPECIFYING AND MONITORING GUARANTEES IN COMMERCIAL GRIDS THROUGH SLA Sven Graupner Vijay MachirajuAad van Moorsel IEEE/ACM International Symposium on Clustering.
MTA SZTAKI Hungarian Academy of Sciences Grid Computing Course Porto, January Introduction to Grid portals Gergely Sipos
Seminar Grid Computing ‘05 Hui Li Sep 19, Overview Brief Introduction Presentations Projects Remarks.
Seminarium on Component-based Software Engineering Jan Willem Klinkenberg CORBA.
USING THE GLOBUS TOOLKIT This summary by: Asad Samar / CALTECH/CMS Ben Segal / CERN-IT FULL INFO AT:
Globus Toolkit 4 hands-on Gergely Sipos, Gábor Kecskeméti MTA SZTAKI
1-2.1 Grid computing infrastructure software Brief introduction to Globus © 2010 B. Wilkinson/Clayton Ferner. Spring 2010 Grid computing course. Modification.
Milos Kobliha Alejandro Cimadevilla Luis de Alba Parallel Computing Seminar GROUP 12.
Grids and Grid Technologies for Wide-Area Distributed Computing Mark Baker, Rajkumar Buyya and Domenico Laforenza.
4b.1 Grid Computing Software Components of Globus 4.0 ITCS 4010 Grid Computing, 2005, UNC-Charlotte, B. Wilkinson, slides 4b.
Mike Smorul Saurabh Channan Digital Preservation and Archiving at the Institute for Advanced Computer Studies University of Maryland, College Park.
UMIACS PAWN, LPE, and GRASP data grids Mike Smorul.
The Open Grid Service Architecture (OGSA) Standard for Grid Computing Prepared by: Haoliang Robin Yu.
Globus Computing Infrustructure Software Globus Toolkit 11-2.
OGSA : Open Grid Services Architecture Ramya Rajagopalan
Web-based Portal for Discovery, Retrieval and Visualization of Earth Science Datasets in Grid Environment Zhenping (Jane) Liu.
Globus 4 Guy Warner NeSC Training.
Kate Keahey Argonne National Laboratory University of Chicago Globus Toolkit® 4: from common Grid protocols to virtualization.
Data Management Kelly Clynes Caitlin Minteer. Agenda Globus Toolkit Basic Data Management Systems Overview of Data Management Data Movement Grid FTP Reliable.
OPEN GRID SERVICES ARCHITECTURE AND GLOBUS TOOLKIT 4
An XMPP (Extensible Message and Presence Protocol) based implementation for NHIN Direct 1.
SOS EGEE ‘06 GGF Security Auditing Service: Draft Architecture Brian Tierney Dan Gunter Lawrence Berkeley National Laboratory Marty Humphrey University.
1.View Description 2.Primary Presentation 3.Element Catalog Elements and Their Properties Relations and Their Properties Element Interfaces Element Behavior.
DISTRIBUTED COMPUTING
Software Architecture Framework for Ubiquitous Computing Divya ChanneGowda Athrey Joshi.
ANSTO E-Science workshop Romain Quilici University of Sydney CIMA CIMA Instrument Remote Control Instrument Remote Control Integration with GridSphere.
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
Grid Security Issues Shelestov Andrii Space Research Institute NASU-NSAU, Ukraine.
Grid Resource Allocation and Management (GRAM) Execution management Execution management –Deployment, scheduling and monitoring Community Scheduler Framework.
XMPP Concrete Implementation Updates: 1. Why XMPP 2 »XMPP protocol provides capabilities that allows realization of the NHIN Direct. Simple – Built on.
The Grid System Design Liu Xiangrui Beijing Institute of Technology.
The Anatomy of the Grid Mahdi Hamzeh Fall 2005 Class Presentation for the Parallel Processing Course. All figures and data are copyrights of their respective.
Virtual Data Grid Architecture Ewa Deelman, Ian Foster, Carl Kesselman, Miron Livny.
The Anatomy of the Grid Introduction The Nature of Grid Architecture Grid Architecture Description Grid Architecture in Practice Relationships with Other.
Middleware for Grid Computing and the relationship to Middleware at large ECE 1770 : Middleware Systems By: Sepehr (Sep) Seyedi Date: Thurs. January 23,
Grid Architecture William E. Johnston Lawrence Berkeley National Lab and NASA Ames Research Center (These slides are available at grid.lbl.gov/~wej/Grids)
Ames Research CenterDivision 1 Information Power Grid (IPG) Overview Anthony Lisotta Computer Sciences Corporation NASA Ames May 2,
NA-MIC National Alliance for Medical Image Computing UCSD: Engineering Core 2 Portal and Grid Infrastructure.
GRID Overview Internet2 Member Meeting Spring 2003 Sandra Redman Information Technology and Systems Center and Information Technology Research Center National.
Prof S.Ramachandram Dept of CSE,UCE Osmania University
1 Observations on Architecture, Protocols, Services, APIs, SDKs, and the Role of the Grid Forum Ian Foster Carl Kesselman Steven Tuecke.
1 BBN Technologies Quality Objects (QuO): Adaptive Management and Control Middleware for End-to-End QoS Craig Rodrigues, Joseph P. Loyall, Richard E. Schantz.
1 G52IWS: Web Services Chris Greenhalgh. 2 Contents The World Wide Web Web Services example scenario Motivations Basic Operational Model Supporting standards.
Ad Hoc VO Akylbek Zhumabayev Images. Node Discovery vs. Registration VO Node Resource User discover register Resource.
Globus and PlanetLab Resource Management Solutions Compared M. Ripeanu, M. Bowman, J. Chase, I. Foster, M. Milenkovic Presented by Dionysis Logothetis.
Introduction to Grid Computing and its components.
GRID ANATOMY Advanced Computing Concepts – Dr. Emmanuel Pilli.
1 AHM, 2–4 Sept 2003 e-Science Centre GRID Authorization Framework for CCLRC Data Portal Ananta Manandhar.
Super Computing 2000 DOE SCIENCE ON THE GRID Storage Resource Management For the Earth Science Grid Scientific Data Management Research Group NERSC, LBNL.
EGI-InSPIRE RI EGI-InSPIRE EGI-InSPIRE RI EGI Services for Distributed e-Infrastructure Access Tiziana Ferrari on behalf.
The EPIKH Project (Exchange Programme to advance e-Infrastructure Know-How) gLite Grid Introduction Salma Saber Electronic.
Accessing the VI-SEEM infrastructure
The Open Grid Service Architecture (OGSA) Standard for Grid Computing
GGF OGSA-WG, Data Use Cases Peter Kunszt Middleware Activity, Data Management Cluster EGEE is a project funded by the European.
University of Technology
Core Grid Functions: A Minimal Architecture for Grids
Core Grid Functions: A Minimal Architecture for Grids
Grid Services B.Ramamurthy 12/28/2018 B.Ramamurthy.
Introduction to Grid Technology
Enterprise Integration
The Anatomy and The Physiology of the Grid
The Anatomy and The Physiology of the Grid
Presentation transcript:

Core Grid Functions: A Minimal Architecture for Grids William E. Johnston Lawrence Berkeley National Lab and NASA Ames Research Center (www-itg.lbl.gov/~wej) Work of the GGF Grid Protocol Architecture WG W. Johnston, J. Brooke, white paper co-authors

2 Core Grid Functions Goal: –A minimal set of Grid functions that provide uniform interfaces and management for architecturally, geographically, and administratively heterogeneous computing, data, and instrument systems –that are managed as production Grids “Production Grids” are the Grids that are trying to provide services to a diverse user community to whom the operators of the Grid are responsible for providing a reliable and useful service Note: Interoperability also requires operational agreements

3 Core Grid Functions This minimal set of functions are the smallest set of services that are needed to build all other Grid frameworks, middleware, and applications –the minimal services may vary somewhat depending on the type of Grid resource – computing, data, instrument, etc.

4 Core Grid Functions Defining a “minimal” set of functions is important because: –They provide a metric related to whether a system is a Grid enabled system, or not –without the Core Grid Functions, there will be Grid middleware, frameworks, and applications that cannot function –represent the fundamental persistent infrastructure of the Grid –represent most of the operational effort in building and managing Grids

5 Core Grid Functions Criteria for a Core Grid Function –cannot be built on top of other Grid services –is essential for building other Grid services and applications, or for providing scalability or security –must be self contained (except possibly with respect to security)

pools of work- stations clusters national super- computer facilities... Encapsulation as Python Services, Script Based Services, Java Based Services, … Distributed Resources space-based networksoptical networksInternet Communications Portals Higher Level Services (applications, utilities, etc.) Identity Credential Management Grid Security Infrastructure Globus 2-style interface Service discovery Lifecycle management Service registry Service factory (execution) Service handleMap Notification (events) OGSA Core Grid Functions – protocol endpoints tertiary storage scientific instruments Security Gateways information servers J2EE hosting environment servers Factory services Grid Security Infrastructure: Authentication (human, host, service), delegation/proxy, secure communication Uniform Data Access Events, Monitoring, Logging Uniform Computing Access Resource Scheduling Mg’mt Access (remote shell & cpy) Authorization Persistent state and Registry resource characteristics, internal architecture, operating state, dynamic registry event data types Dataset replica info. VO information Execution environment establishment OGSA hosting environment Grid Auxiliary Functions Core Grid Functions – protocol endpoints Authentication and Security Resource discovery Resource Scheduling Events and Monitoring Uniform Computing Access Uniform Data Access Communication Frameworks (Legion-G, CORBA, ….) Proxy servers (NAT, FTP cache, etc.) Mg’mt Access (remote shell & cpy) Brokering Job mg’mt (e.g. Condor-G, Unicore) DataGrid services (e.g. replication and naming) Workflow engine Toolkits and collective services Unix shell hosting environment Configuration based workflow transformation

7 Core Grid Functions Resource Discovery and State / Grid Persistent State Resource Scheduling Uniform Computing Access Uniform Data Access Asynchronous Information Sources (Events, Monitoring, Logging, etc.) Remote Authentication, Authorization, Delegation, and Secure Communication System Management and Access Architectural Constraints (e.g. security) Bindings

8 Resource Discovery & State / Grid Persistent State A Grid information service must provide information about existence and characteristics of all Grid resources Should minimize the number of persistent information servers needed to enable Grid services and applications Functionality –Provide for locating all Grid resources with specified properties, within a certain scoping –Provide state information as pointers –Accommodate a dynamic resource base –Be extensible to “all” Grid persistent state all Grid services can be sources of information, and if this information needs to be referenced and/or discovered, it should be possible to store and/or represent it in the Grid information service. E.g. Data from users, Virtual Organizations, applications Computing resources Available software Current user allocation Asynchronous Information Sources registry and data content

9 Resource Discovery / Grid Persistent State A minimal service –Discovery is an essential Grid function. Without discovery, you cannot build virtual systems from dynamically changing pools of resources. –Management of persistent servers is operationally expensive, therefore it is critical to minimize the number of servers needed for a persistent Grid Storing / representing all manner of persistent Grid information with one service is important to minimize required operational support

10 Resource Scheduling Scheduling coordinates distinct resources so that they may operate cooperatively Functionality –Establish a given virtual system relationship among an administratively independent set of Grid resources via co- scheduling –Return information sufficient for negotiation of a common QoS (e.g. time slot) among independent resources –A scheduler operating on the resource must Provide time of day reservation Evaluate the future availability of a reservation request and pass that information back to the requester Support soft reservations to allow time for an external broker to negotiate a common reservation among several resources A minimal service –Essential for QoS –Not possible to emulate

11 Uniform Computing Access Job / process initiation Functionality –Initiate a process or task script on a remote Grid system –Support queries about queue types –Support submission to named queues (different classes of service) –Perform access control based on Grid identity –Adapt to variations in system architecture

12 Uniform Computing Access Execution environment establishment –Hosting (provide for certain Grid styles of I/O, IPC, etc.) OGSA Unix shell –Establish the application runtime environment –Configuration based workflow transformation

13 Uniform Data Access Today the primary Grid data access is to named, unstructured objects (“flat” files) –objects / files whose structure is understood only by the application that reads the files, and not by the storage system –Hence, the primary current model for Grid data access is FTP Other emerging functionality in Grid storage resources: –Support for some mechanism of sub-setting or filtering data before it leaves the storage resource –providing access to relational databases –providing access to object oriented databases (?) Flat File / Unstructured Object Access Functionality –Storage access abstraction –Partial file access –Integrated Grid security infrastructure security and access control based on the Grid identity

Asynchronous Information Sources (Events, Monitoring, Logging, etc.) “Asynchronous Information Sources” = any source of XML formatted objects that can publish its existence and object content characteristics, and then support subscription based delivery of those objects Functionality –Source registration (a la GMA, the source registers its existence and the content of the objects that it will generate) –Registry should be “globally” searchable based on various source and/or object content characteristics –Receiving data is by subscription and by direct transfer (source to sink) – the GMA model A minimal service –Generally, users cannot start persistent servers –May be required on systems where jobs cannot be initiated (e.g. storage and instrument control systems)

Remote Authentication, Authorization, Delegation, and Secure Communication Identity Certification Authority and certificate management –Provides a mechanism for users / entities to request certificates –Provides a registration process that verifies user/entity identity –Issues and signs X.509 identity certificates –Provides Certificate Revocation List generation, management, access, and use –Provides a certificate repository –Has a formal policy

Remote Authentication, Authorization, Delegation, and Secure Communication Authentication –Authenticate user access based on Grid identity cert –Provide for using host identity credentials at both ends of a transport connection for validating the system identities Securely conveying user entity credentials/proxy to the remote system Authorization –Access control based on Grid identity and attributes Secure Communication –Encrypted streams and messages Delegation –The process by which a user’s identity (perhaps with restrictions) is carried to a remote system without the user being directly involved at the remote system These are all essential components for secure, authenticated, and authorized access to remote systems

17 System Management and Access Remote system management, and sometimes remote user access, are needed so that Grid resources may be managed and interactively accessed within the Grid context Functionality –Remote login, authenticated and secured with Grid security functions and authorization based on Grid identity –Remote shell, authenticated and secured with Grid security functions and authorization based on Grid identity –Remote copy, authenticated and secured with Grid security functions and authorization based on Grid identity This seems to be an essential service, because if it is not provided then it is always accomplished in a ad- hoc manner

18 Architectural Constraints In order to be called a Grid Service, it should not be possible to convey command and control messages to remote Grid systems except through the secure and authenticated communication provided by the Grid security functions Secure data channels should always be optional, as encryption may be impractical in the cases of high data rates or volumes

19 Bindings Most of the Core Functions will be defined in terms of protocols and data structures, and this provides the basic uniformity required of Grids However, there will be many ways to use these Core Functions. For example –Globus toolkit’s C language –CoG kit’s Java interface to the Globus functions –PyGlobus interface to the Globus functions Arguably the OGSI work represents a non- “Globus” interface to the Core functions And there will be others

20 This talk is at grid.lbl.gov/~wej/Grids