GRID COMPUTING
CONTENTS:
Introduction What is Grid Computing Characteristics of Grid Components of Grid Advantages Challenges Grid development option Motivations for Grid Computing How Grid Works Grid Protocol Strategies for Grid Application Enablement Area of Application Summery References
1
Introduction
What is a distributed environment? “A non-centralized network consisting of numerous computers that can communicate with one another and that appear to users as parts of a single, large, accessible "storehouse" of shared hardware, software, and data”
2
Introduction - Single Computer •The Operating System enables easy use of –Input/Output devices –Processor –Disks –Display –Instruments
Application Software Operating System Disks, Processor, Memory, …
3
Introduction - Local Area Network User just perceives “shared resources”, with no regard to location in the organisation LAN resources act like a single virtual computer Middleware (LAN O/S) presents that image
Application Software Middleware for sharing computers, servers, printers, … Operating System on each computer
Resources connected by a LAN
4
Introduction - Grid •Grid middleware creates the image of the Grid being a single virtual computer (Ideally) Issues •Heterogeneity – hardware, software, culture •Scalability •Reliability – tolerate permanent partial failure •Viable computing model batch processing •Access control –Authentication –Authorisation –Single sign on
Application Software Interface between app. and grid Grid Middleware on each resource Operating System on each resource
Resources connected by internet
5
WHAT IS GRID COMPUTING ?
It is the distributed computing taken to the next evolutionary level. The goal is to create the illusion of a simple yet , large and powerful self managing virtual computer out of a large connected heterogeneous system sharing resources.
What is not a Grid?
A bunch of PC’s on a network (It’s a lot more than that) A cluster, a network attached storage device, a network etc. (each is an important component of grid, but by itself each does not constitute a grid)
6
WHAT IS GRID COMPUTING ?
Distributed Computing
Cluster Computing
Loosely Coupled Heterogeneous Single Administrator Tightly Coupled Homogenous Cooperative Working
Grid Computing
Large Scale Cross Organization Geographical Distribution Distributed Management
7
Characteristics of Grid
Co-ordinated resource sharing
Standard, open, general-purpose protocols and interfaces
No centralised point of control Different administrative domains.
NOT specific to an application support multiple VO’s
Delivering non-trivial qualities of service
Co-ordinated to deliver combined services, greater than sum of the individual components
8
Components of Grid •
Resources –
•
Grid Middleware –
•
networking, computers, storage, data, instruments, … the “operating system of the grid”
Virtual Organization management –
Procedures for gaining access to resources
9
Advantages :
Virtualized Sharing of Resources Secure reliable access to Resources Autonomic management of Resources Proper Utilization of Resources Fast Computation (nearly achievable to Super Computing by Parallel Computing) Virtually a very Large Capacity Economic No need of nodes homogeneity
10
Challenges
Non-determinism Infrastructure dependencies Distributed and partial failures Time-outs Dynamic nature of the structure Multiple heterogeneous platforms Security The ‘Human Factor’
11
Grid development option
A function of Business need, Technology and Organizational flexibility
12
Grid development option
A function of Business need, Technology and Organizational flexibility
13
Grid development option
A function of Business need, Technology and Organizational flexibility
14
Motivations for Grid Computing
15
Motivations for Grid Computing
Increase Capacity Exploits distributed recourses to provide capacity for high demand applications
16
Motivations for Grid Computing
Increase Capacity Exploits distributed resources to produce capacity for high demand applications
Improve Efficiency /Reduce Cost
17
Motivations for Grid Computing
Provide Reliability / Availability
18
Motivations for Grid Computing
Reduce “Time to Results”
19
Motivations for Grid Computing
Provide Reliability / Availability
Use Distributed processes Monitor Work Process Restart failed jobs
Support Heterogeneous system
20
Motivations for Grid Computing
Enable collaborations
21
How Grid Works? 1. 2. 3. 4.
Resource Sharing Resource Utilization Secure Access The Death of Distance
Resource Sharing
1.
CPU Consideration Data Consideration
22
CPU – Makes Execution Parallel Rearranging Computations to execute in Parallel on Grid
23
CPU – Programming Code Control Graph
24
How Grid Works? 2. Resource Utilization
25
How Grid Works? 3. Secure Access
Authentication Authorization Data integrity Data confidentiality Key management encryption
CA - Certification Authority The role of the CA is manage the certificate life cycle: create, store, renew. 26
CA - Certification Authority
It depends on person whether to participate in Grid
Virtual Community C
Person B (Administrator) Compute Server C1'
Person A (Principal Investigator)
Compute Server C1 Person A (Faculty) Person C (Student)
Organization A
Person E (Researcher)
Person D (Researcher)
Person B (Staff) Compute Server C2
File server F1 (disk A)
Person D File server F1 (Staff) (disks A and B)
Person E (Faculty) Person F (Faculty)
Compute Server C3
Organization B
27
CA - Certification Authority 1. Positively identify entities requesting certificates 2. Issuing, removing, and archiving certificates 3. Protecting the Certificate Authority server 4. Maintaining a namespace of unique names for certificate owners 5.Serve signed certificates to those needing to authenticate entities
Digital Signature
28
How Grid Works?
Grid Security
Site A
Site B
Grid Infrastructure R
R
R
Grid Site Security
Security Site C
29
How Grid Works? 4. The Death of Distance development of networking technology Internet economy optical fibers in telecommunications systems performance of wide area networks has been doubling every nine months
30
Grid Protocols 1. 2. 3. 4. 5.
Fabric Connectivity Resource Collective Application
31
Grid Protocol v/s Internet Protocol
32
Grid Protocol - Fabric
33
Grid Protocol - Connectivity
34
Grid Protocol - Resource
35
Grid Protocol - Collective
36
Grid Protocol – Application Layer
37
Six Strategies for Grid Application Enablement 1. 2. 3. 4. 5. 6.
Batch Anywhere Independent Concurrent Batch Parallel Batch Service Parallel Services Tightly Coupled parallel Program
38
Six Strategies for Grid Application Enablement 1. Batch Anywhere Only Grid decides which node to use for the job. The machine submitting the job might not be a node in the grid. 2. Independent Concurrent Batch Multiple independent instances of the same application run concurrently and independently without interface. Databases and other resources don’t have deadlocks. 3. Parallel Batch Take each user’s batch work, subdivides it, disperse it out to multiple nodes, collect it, and then aggregate the result. 4. Service Focuses on the transition from a batch to service oriented architecture. It is not assumed that each client subdivides it’s work and spreads it over multiple service instances. 5. Parallel Services Service with subdivided work model of parallel batch. Provides multiple service instance Permits these instances to be invoked in parallel on the client’s behalf. 6. Tightly Coupled Parallel Programs Provides intense communications and synchronization between client, server and services.
39
Area of Application
40
Area of Implementation
41
Grid Computing – Industry Applications Unique by Industry with Common Characteristics
Primary Focus
42
Grid Project in China (river Vah)
43
Data Intensive Applications
44
Data Intensive Application Overlapping Data Transfer with Capture and Computing
45
Many Grid development efforts — all over the world •NASA Information Power Grid •DOE Science Grid •NSF National Virtual Observatory •NSF GriPhyN •DOE Particle Physics Data Grid •NSF TeraGrid •DOE ASCI Grid •DOE Earth Systems Grid •DataGrid (CERN, ...) •DARPA CoABS Grid •EuroGrid (Unicore) •NEESGrid •DataTag (CERN,…) •DOH BIRN •Astrophysical Virtual •NSF iVDGL Observatory
•UK – OGSA-DAI, RealityGrid, GeoDise, Comb-e-Chem, DiscoveryNet, DAME, AstroGrid, GridPP, MyGrid, GOLD, eDiamond, Integrative Biology, … •Netherlands – VLAM, PolderGrid •Germany – UNICORE, Grid proposal •France – Grid funding approved •Italy – INFN Grid •Eire – Grid proposals •Switzerland - Network/Grid proposal •Hungary – DemoGrid, Grid proposal •Norway, Sweden - NorduGrid
•GRIP (Globus/Unicore) •GRIA (Industrial applications) •GridLab (Cactus Toolkit) •CrossGrid (Infrastructure Components) •EGSO (Solar Physics)
46
Can Your Application Benefit from “Grid Computing”
47
CONCLUSION The grid will bring about a revolution in the way we use our resources. Twenty years and so, the grid will no longer be a domain of only researchers but a tangible entity that shall be instinct to the way we live. Uncouthly, grid will be the next generation Internet.
48
References
“The Anatomy of Grid” by Foster and Kesselman http://www.globus.org http://www.ibm.com/redbooks/ http://www.gridtoday.com http://www.redbooks.ibm.com http://www.google.com
49
Thank You
Questions?