the anatomy of the grid mahdi hamzeh fall 2005 class presentation for the parallel processing...
TRANSCRIPT
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 authors.
Outline
Introduction Common requirements Our objectives The nature of grid architecture Relationship with other technologies Other perspective on grids References
Introduction
The Grid Coined in the mid-1990s Group of participants Varying degrees of prior relationship Mutually Distrustful Need to share resources to perform a task
Direct access to software, data, sensors and computers Dynamic sharing relationships Sharing subject to a set of constraints
What, Who, When etc. Industry Science Engineering
Our purpose Develop a detailed architecture Roadmap
Introduction (contd)
Real and specific problems Coordinated resource sharing Problem solving in dynamic multi-institutional virtual organization
Why carefully study underlying technology? Common concerns Requirements
Why not current distributed computing technologies is enough? Does not accommodate the range of resource types Does not provide the flexibility Does not provide control on sharing relationships
Common requirements
Highly flexible sharing relationships Ranging
Sophisticated and precise levels of control Sharing of varied resources
Programs Files Data Computers Sensors Networks
Diverse usage modes Single user to multi-user Performance sensitive to cost sensitive Quality of service Accounting
Our objectives
Clarify the nature of VOs and Grid computing
Contribute to the emergence of Grid computing
Define clearly how Grid technologies relate to other technologies
The nature of grid architecture Interoperability
In a networked environment, interoperability means common protocols VO users and resources negotiate, establish, manage, and exploit sharing relationships Accommodating new participants dynamically
Protocols Specifies how distributed system elements interact with one another Structure of the information exchanged during this interaction Discover resources Establish identity Determine authorization Initiate sharing All must flexible and lightweight
Services access to computation access to data resource discovery
Application programming interfaces and software development Programming abstractions
enable code sharing enhance application portability
GRID ARCHITECTURE
Our goal Identify requirements for general classes of
componentOpen architectural structureExtensible
Hourglass model Definition of core abstraction and protocols
Foster, I., Kesselman, C. and Tuecke, S. The Anatomy of the Grid: Enabling Scalable Virtual Organizations. Intl. J. Supercomputer Applications, 2001
The layered Grid architecture and its relationship to the Internet protocol architecture
Foster, I., Kesselman, C. and Tuecke, S. The Anatomy of the Grid: Enabling Scalable Virtual Organizations. Intl. J. Supercomputer Applications, 2001
Fabric
The Grid Fabric layer provides the resources
OperationsEnquiryResource management
ResourcesComputational resourcesStorage resourcesNetwork resources ...
Connectivity
Communication Transport Routing Naming
Authentication Single sign-on Delegation Integration with various local security solutions
Kerberos Unix security
User-based trust relationships
Resource
Information protocols Structure State
Management protocols Negotiation Monitoring Initiation Control Accounting Payment Serve ‘requested protocol operations are consistent with the
policy under which the resource is to be shared’
Collective
Sharing behaviors Directory services coallocation-allocation, scheduling, and brokering services Monitoring and diagnostics services Data replication services Grid-enabled programming systems Workload management systems and collaboration frameworks Software discovery services Community authorization servers Community accounting and payment services Collaboratory services
Example of Implementation
Foster, I., Kesselman, C. and Tuecke, S. The Anatomy of the Grid: Enabling Scalable Virtual Organizations. Intl. J. Supercomputer Applications, 2001
Application
Foster, I., Kesselman, C. and Tuecke, S. The Anatomy of the Grid: Enabling Scalable Virtual Organizations. Intl. J. Supercomputer Applications, 2001
Architecture in practice
Storage systems, computers, networks, code repositories, catalogs
Fabric
Communication (IP), service discovery (DNS), authentication, authorization, delegation
Connectivity
Access to computation; access to data; access to information about system structure, state, performance.
Resource
Resource discovery, resource brokering, system monitoring ,community authorization, certificate revocation
Collective (generic)
Checkpointing , job management, failover, staging
Collective
(application-specific)
Ray Tracing
Solver coupler ,distributed data archiver
Multidisciplinary Simulation
Intergrid protocols
Select and achieve widespread deployment of one set of protocols at the Connectivity and Resource layers
Lesser extent, at the Collective layer
Relationship with other technologies World Wide Web
QoS Guarantees [ No ] Distributed Resources [ Yes ] De-centralized Coordination [ No ] Standard/Open Protocols [ Yes ]
Application and storage service providers QoS Guarantees [ Yes ] Distributed Resources [ Yes/No ] De-centralized Coordination [ No ] Standard/Open Protocols [ No ]
Internet and peer-to-peer computing QoS Guarantees [ No /No] Distributed Resources [ Yes ] De-centralized Coordination [ Yes ] Standard/Open Protocols [ No ]
Other perspective on grids
The Grid is a next-generation Internet The Grid is a source of free cycles The Grid requires a distributed operating system The Grid requires new programming models The Grid makes high-performance computers
superfluous
References Berman, F. Fox, G. Hey, T. Grid Computing making the global infrastructure a reality ,
WILEY,2003. Foster,I.Grid Technologies & Applications: Architecture & Achievements , 2002. Foster, I., Kesselman, C. and Tuecke, S. The Anatomy of the Grid: Enabling Scalable
Virtual Organizations. Intl. J. Supercomputer Applications, (to appear). 2001. Frey, J., Tannenbaum, T., Foster, I., Livny, M. and Tuecke, S., Condor-G: A Computation
Management Agent for Multi-Institutional Grids. In 10th International Symposium on High Performance Distributed Computing, IEEE Press, 2001.
Livny, M. High-Throughput Resource Management. In Foster, I. and Kesselman, C. eds. The Grid: Blueprint for a New Computing Infrastructure, Morgan Kaufmann, 1999
Stockinger, H., Samar, A., Allcock, W., Foster, I., Holtman, K. and Tierney, B., File and Object Replication in Data Grids. In IEEE Intl. Symp. on High Performance Distributed Computing, IEEE Press, 2001.
Armstrong, R., Gannon, D., Geist, A., Keahey, K., Kohn, S.,McInnes, L. and Parker, S.Toward a Common Component Architecture for High Performance Scientific Computing. In Proc. 8th IEEE Symp. on High Performance Distributed Computing,1999.
http://www.globus.org/alliance/publications/papers.php