Anatomy of Grid enabling scalable Virtual Organizations

1
Anatomy of Gridenabling scalable Virtual
Organizations

• Authors
• - Ian Foster, Carl Kesselman,
• Steven Tuecke
• Presented by Nihal Desai

2
Topics

• What is Grid???
• Grid problem
• clarify the nature of Grid computing and VOs
• Identify the principal functions required to
  enable sharing within VOs
• requirements and a framework for a Grid
• Define how Grid technologies relate to other
  technologies
• Strength weakness

3
What is grid???

• A Grid is a system that
• coordinates resources that are not subject to
  centralized control
• using standard, open, general-purpose protocols
  and interfaces
• to deliver nontrivial qualities of service.

4
Example

• A cluster management system such as Suns Sun
  Grid Engine, Platforms Load Sharing facility, or
  Veridians Portable Batch System ???
• the Web ???

5
Differences between Grids and the older
distributed OS

• site autonomy
• heterogeneity
• involve more resources
• focus on the user

6
Grid Computing

• Broken down into two sects
• Computationally focused grids (a.k.a.
  Computational Grids)
• Data-centric grids (a.k.a. Data Grids)

7
Grid Problem

• Grid computing is concerned with coordinated
  resource sharing and problem solving in dynamic,
  multi-institutional virtual organizations.
• Sharing not only file transfer but direct
  access to computers, software, data, and other
  resources. Necessarily, highly controlled, with
  resource providers and consumers defining clearly
  and carefully

8
Virtual Organization

• virtual organization
• A set of individuals and/or institutions
  defined by such sharing rules
• In other words, VOs are dynamic
  federations of heterogeneous organizational
  entities sharing data, metadata, processing and
  security infrastructure
• VOs vary tremendously,but underlying technology
  requirements leads us to identify a broad set of
  common concerns and requirements.

9
Continued

• highly flexible sharing relationships
• sophisticated and precise levels of control over
  how shared resources are used
• sharing of varied resources
• diverse usage modes
• current technology either does not accommodate
  the range of resource types or does not provide
  the flexibility and control on sharing
  relationships needed to establish VOs.

10
Continued
11
Resource sharing properties

• Resource sharing is conditional
• Sharing relationships can vary dynamically over
  time
• Sharing relationships may be combined to
  coordinate use across many resources
• The same resource may be used in different ways

12
The nature of Grid Architecture

• Interoperability is the central issue
• Why??
• Protocol Architecture services
• Why??
• API and SDK
• Why??

13
Grid Architecture

• Hourglass Model
• In our architecture, the neck of the
  hourglass consists of Resource and Connectivity
  protocols which facilitate the sharing of
  individual resources
• Protocols are designed so that they can be
  implemented on top of a diverse range of resource
  types, defined at the Fabric layer
• can in turn be used to construct a wide range of
  global services and application-specific
  behaviors at the Collective layer

14
Five layered Grid Architecture
15
Layers
16
Fabric LayerInterface to local control

• Abstraction of underlying systems, data,
  resources
• Fabric components implement the local,
  resource-specific operations that occur on
  specific resources (whether physical or logical)
  as a result of sharing operations at higher
  levels
• If we place few demands on Fabric elements, then
  deployment of Grid infrastructure is simplified
  

17
Fabric Layer continuedResource mechanisms

• Minimum Inquiry resource management mechanism
• Computational resources- start/end/monitor
  control
• Storage resources- putting and getting files
• Network resources-managing network transfer
• Code repositories-managing versioned codes
• Catalogs- Query and Update

18
Globus Toolkit

• designed to use existing fabric components
• if a vendor does not provide the necessary
  Fabric-level behavior, the Globus Toolkit
  includes the missing functionality
• Example

19
Connectivity Layer communicating easily/securely

• This is where the security comes in. defines core
  communication and authentication protocols
  required for Grid-specific network transactions
• Communication transport, routing, and naming
  based on TCP/IP stack
• Authentication Single sign on, Delegation,
  Integration with various local security
  solutions, User-based trust relationships

20
Globus Toolkit

• The Internet protocols are used for communication
• (GSI) protocols are used for authentication,
  communication protection, and authorization
• GSI builds on and extends the Transport Layer
  Security (TLS) protocols to address delegation,
  integration with various local security solutions
  and user-based trust relationships.

21
Resource Layer sharing single resources

• these protocols call Fabric layer functions to
  access and control local resources. secure
  negotiation, initiation, monitoring, control,
  accounting payment of sharing operations on
  individual resources.
• Manages and provides APIs SDKs to each
  available grid resources.
• concerned entirely with individual resources
• Two primary classes of Resource layer protocols
• Information protocols Management
  protocols

22
Globus Toolkit

• GRIP
• GRRP
• GRAM
• GridFTP
• LDAP as a catalog access protocol.

23

• Resource and Connectivity protocol layers form
  the neck of our hourglass model, and as such
  should be limited to a small and focused set.
• These protocols must be chosen so as to capture
  the fundamental mechanisms of sharing across many
  different resource types (for example, different
  local resource management systems)

24
Collective Layer coordinating multiple resources

• Components that deal with the coordination and
  cooperation of many grid resources
• global in nature and capture interactions across
  collections of resources.
• Collective layer protocols span the spectrum from
  general purpose to highly application or domain
  specific
• Collective functions can be implemented as
  persistent services, with associated protocols,
  or as SDKs (with associated APIs) designed to be
  linked with applications.

25
Continued
26
Continued

• Directory services
• Co-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 accounting and payment services
• Collaboratory services

27
Globus Toolkit

• In addition to other services, Meta Directory
  Service which introduces Grid Information Index
  Servers (GIISs) to support arbitrary views on
  resource subsets
• replica catalog and replica management services
  to support the management of dataset replicas
• online credential repository service (MyProxy)
  for secure storage for proxy credentials
• DUROC co-allocation library provides an SDK and
  API for resource co allocation

28
Application Layer

• This is where you use publicly available APIs to
  interface with underlying infrastructure
• architecture comprises the user applications that
  operate within a VO environment

29
Application programmers view
30
Grid Architecture in practice
31
Relationships with Other Technologies

• Current distributed computing approaches do not
  provide a general resource-sharing framework that
  addresses VO requirements
• Because of their focus on dynamic,
  cross-organizational sharing, Grid technologies
  complement rather than compete with existing
  distributed computing technologies
• WWW
• Application and Storage Service
  Providers
• Enterprise Computing Systems
• Internet and Peer-to-Peer Computing

32
WWW

• lack features required for the richer interaction
  models that occur in VOs.
• Example
• use TLS for authentication, but do not support
  single sign-on or delegation.
• the single sign-on capabilities provided in the
  GSI extensions to TLS would allow for single
  sign-on to multiple Web servers

33
Application storage service providers

• VPNs and static configurations make many VO
  sharing modalities hard to achieve
• The integration of Grid technologies into ASPs
  and SSPs can enable a much richer range of
  possibilities.

34
Enterprise Computing Systems

• Sharing arrangements are typically relatively
  static and restricted to occur within a single
  organization.
• Example
• in the case of CORBA, we could construct an
  object request broker (ORB) that uses GSI
  mechanisms to address cross-organizational
  security issues.

35
Internet peer-to-peer computing

• Lacking common protocols that would allow for
  shared infrastructure interoperability
• sharing targeted by various applications are
  quite limited

36
Other Perspectives

• 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.

37
Strength Weakness

• Perfect paper for Grid VO fundamentals
• Defines what to be done and not how to be done
• Emphasis on Globus Toolkit, not on other
  technologies

38
Questions???