SCC717 Recent Developments in Information Technology Lecturer: Wanlei ZHOU, Ph.D., DSc., Professor, Acting Head School of Information Technology Deakin University, Melbourne Campus Office: T205.2, Ext. 17603 URL: http://www.cm.deakin.edu.au/~wanlei Email: wanlei@deakin.edu.au
! 2.1 Study Points! Understand the evolution of grid computing.! Identify the main challenges for grid computing.! Be familiar with the technologies that support grid computing.! Identify the main applications of grid computing
! 2.2 Grid Computing! Evolution of Network Computing: to eliminate resource islands (Prof. K. Huang). Grid Applications Grid:// Applications Web pages WWW http://... Web pages Computers Internet ftp://... telnet://... Computers
GGG(Great Global Grid), Forbes 2001 Computing and Data Grids Information and Knowledge Grids Business Grids Other Grid Models TeraGrid Vega Grid ASCI Grid Data Grid Semantic Web Knowledge Management Ontology Platform HP espeak/mmgrid IBM WS/Grid Toolbox Microsoft.NET/ Internet OS Sun ONE Grid Engine P2P Parasitic Computing Desktop Grid (Entropia)
! What is Grid Computing?! Grid Computing aims to couple geographically distributed resources and offer transparent, powerful, and inexpensive services irrespective of the physical location of resources or access points.! Challenges:! To seamlessly integrate geographically distributed resources, such as computing devices, clusters, data storage, and networks;! To provide a high available and high reliable source of computing power;! To automatically allocate resources according to user preferences and computational demand;! To provide sophisticated analysis, debugging and visualisation services, and! To facilitate and enable information and knowledge sharing.
! Cluster and Grid:! A Cluster requires a Single System Image (SSI) while a Grid does not require a single system view (although is could have SSI).! A Cluster emphasises on performance of parallel processing while a Grid emphasises on resource sharing.! Geographically a Cluster is located in a central location while a Grid is distributed in many places.! Normally computers in a cluster are homogenous, while a grid involves heterogeneous computers and other resources.
! Basic functions needed for a grid application! Discovery and brokering.! Metering and accounting.! Data sharing.! Resource management.! Security.! Reliability and availability.! Virtual organizations.! Monitoring.! Policy implementation and enforcement.
! The Open Grid Services Architecture (OGSA) Platform! To define standard approaches to, and mechanisms for, basic problems that are common to a wide variety of Grid systems, such as communicating with other services, establishing identity, negotiating authorization, service discovery, error notification, and managing service collections.! The three principal elements of the OGSA Platform are the Open Grid, Services Infrastructure, OGSA Platform Interfaces, and OGSA Platform Models.! Building on both Grid and Web services technologies, the Open Grid Services Infrastructure (OGSI) defines mechanisms for creating, managing, and exchanging information among entities called Grid services.! OGSA Platform Interfaces build on OGSI mechanisms to define interfaces and associated behaviors for various functions not supported directly within OGSI, such as service discovery, data access, data integration, messaging, and monitoring.! OGSA Platform Models support these interface specifications by defining models for common resource and service types.
1. Foster, I., Kesselman, C., Nick, J. and Tuecke, S. The Physiology of the Grid: An Open Grid Services Architecture for Distributed Systems Integration, Globus Project, 2002. www.globus.org/research/papers/ogsa.pdf 2. I. Foster and D. Gannon, Open Grid Services Architecture Platform, 2003, http://www.ggf.org/ogsa- wg
! Key Features of the Open Grid Services Infrastructure! Grid service descriptions and instances. Grid service descriptions define interfaces and behaviors, and Grid service instances implement those behaviors! Mechanisms for representing and accessing service state and metadata from a service instance! Naming and name resolution! Fault model: a common approach for conveying fault information from operations! Mechanisms for managing the lifecycle of a Grid service instance! A means of organizing groups of service instances
! Key Features of OGSA Platform Interfaces! Name resolution and discovery! Service domains! Security! Policy! Data management! Messaging, queuing, and logging! Events! Metering and accounting! Transactions! Service orchestration! Administration! Provisioning and resource management.
! 2.3 Readers and Selected References! Grid Services for Distributed System Integration, by I. Foster, C. Kesselman, J. M. Nick, and S. Tuecke, in IEEE Computer, June 2002, pp. 37-46.! Security Implications of Typical Grid Computing Usage Scenarios, by M. Humphrey and M. R. Thompson, in Proceedings of the 10th IEEE International Symposium on High Performance Distributed Computing (HPDC-10 01), 2001.! http://www.gridcomputing.com/. The Grid Computing Web site maintained by Dr Rajkumar Buyya in Melbourne University.
! 2.4 Review Questions 1. What is the evolution path of grid computing so far and what would be its future for the next 10 to 20 years? 2. How do you differentiate cluster and grid? 3. What are the major projects of grid computing and what are their main characteristics? 4. What are the major challenges of grip computing and why? 5. Grid computing can be viewd as one of the major information technology infrastructures of a nation. Why? 6. Decribe the major characteristics of a computational grid and the applications that may use such a grid. 7. Describe and differentiate a data grid, an information grid, and a knowledge grid. 8. Why trust is a major issue in grid computing and what are the current technologies / methods to implement the trust?