GRID ARCHITECTURE

Grid Architecture

Grid architecture refers to those aspects of a grid system that are taken into consideration when a grid is designed and implemented. Grid architecture can be visualized as a layered architecture

 

§  GRID ARCHITECTURE

o   identifies fundamental system components,

o   specifies the purpose and function of these components, and

o   indicates how these components interact with one another.

 

Fabric: Interfaces to Local Control

§  Function – controls access to shared resources

§  includes the protocols and interfaces that provide controlled access to shared resources, including computational resources, storage systems, datasets, programs, sensors, networks...

§  resource may also be a logical entity, such as a distributed file system, computer cluster, or distributed computer pool;

§  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.

§  Resources implement

o   introspection mechanisms that permit discovery of their structure, state, and capabilities (e.g., whether they support advance reservation), and

o   resource management mechanisms that provide some control of delivered quality of service.

Shared Resources

§  Computational resources.

o   Mechanisms are required for starting programs and for monitoring and controlling the execution of the resulting processes.

o   Management mechanisms that allow control over the resources allocated to processes are useful, as are advance reservation mechanisms.

§  Introspection functions – determining hardware and software characteristics as well as relevant state information such as current load and queue state in the case of scheduler-managed resources.

 

§  Storage resources.

o   Mechanisms are required for getting / storing files.

o   reading and writing subsets of a file and/or

o   executing remote data selection or reduction functions (118).

o   Management mechanisms –

§  allow control over the resources allocated to data transfers (space, disk bandwidth, network bandwidth, CPU) are useful,

§  advance reservation mechanisms.

§  Introspection functions are needed for determining hardware and software characteristics as well as relevant load information such as available space and bandwidth utilization.

§  Network resources.

o   Management mechanisms

§  provide control over the resources allocated to network transfers (e.g., prioritization, reservation)

§  Introspection functions should be provided to determine network characteristics and load.

o   Other important classes of resources include database systems used to store structured data and sensors of various kinds.

 

Connectivity Layer: Communicating Easily and Securely

 

§  The connectivity layer defines core communication protocols and Security protocols (authentication and authorization) required for Grid-specific network transactions.

§  Communication requirements include transport, routing, and naming.

§  Communication protocols enable the exchange of data between fabric layer resources.

§  Authentication protocols build on communication services to provide cryptographically secure mechanisms for verifying the identity of users and resources.

 

§  Security requirements:

o   Single sign-on. As Grid users frequently want to initiate computations that access multiple remote resources, a user should be able to “sign on” (authenticate) just once, rather than once per resource or administrative domain accessed.

o   Delegation. A user must be able to endow a program with the ability to run on the user’s behalf, so that the program is able to access the resources on which the user is authorized. The program should (optionally) also be able to delegate a subset of its rights to another program: what is sometimes referred to as restricted delegation.

§  Integration with local security solutions. In a heterogeneous Grid, each site or resource provider may employ any of a variety of local security solutions.

§  User-based trust relationships. For a user to use resources from multiple providers together, the security system must not require each of the resource providers to cooperate or interact with each other in configuring the security environment. For example, if a user has the right to use sites A and B, the user should be able to use sites A and B together without requiring that A’s and B’s security administrators interact.

 

 

Resource Layer: Sharing Single Resources

 

§  Resource layer defines protocols to initiate and control sharing of (local) resources.  Services defined at this level are gatekeeper, GRIS, along with some user oriented application protocols from the Internet protocol suite, such as file-transfer.

o   (Grid Resource Information Service is the repository of local resource information derived from information providers)

§  The Grid user also needs to be able to interact with remote resources and services. The resource layer, builds on connectivity layer communication and authentication protocols to define protocols for the secure negotiation, initiation, monitoring, control, accounting, and payment of sharing operations on individual resources. Resource layer implementations of these protocols call on fabric layer functions to access and control local resources.

§  Two primary classes of resource layer protocols can be distinguished:

o   Information protocols are used to obtain information about the structure and state of a resource, for example, its configuration, current load, and usage policy (e.g., cost).

o   Management protocols are used to negotiate access to a shared resource, specifying, for example, resource requirements (including advanced reservation and quality of service) and the operation(s) to be performed, such as process creation or data access.

 

Collective: Coordinating Multiple Resources

 

§  Collective layer defines protocols  that provide system oriented capabilities that are expected to be wide scale in deployment and generic in function.  This includes GIIS,  bandwidth brokers and resource brokers

o   GIIS – Grid Index Information Service: represents a centralized MDS server that provides information about all of your resources

o   MDS – Master Data Services (MDS) enables your organization to manage a trusted version of data

 

§  Collective components implement a wide variety of sharing behaviors without placing new requirements on the resources being shared. These include:

o   Directory services allow VO participants to discover the existence and/or properties of VO resources.

§  A directory service may allow its users to query for resources by name and/or by attributes such as type, availability, or load

o   Coallocation, scheduling, and brokering services allow VO participants to request the allocation of one or more resources for a specific purpose and the scheduling of tasks on the appropriate resources.

o   Monitoring and diagnostics services support the monitoring of VO resources for failure, adversarial attack (“intrusion detection”), and overload.

o   Data replication services support the management of VO storage (and perhaps also network and computing) resources to maximize data access performance with respect to metrics such as response time, reliability, and cost.

o   Grid-enabled programming systems enable familiar programming models to be used in Grid environments, using various Grid services to address resource discovery, security, resource allocation, and other concerns. Examples include Grid-enabled implementations of the Message Passing Interface and manager–worker frameworks.

o   Workflow systems provide for the description, use, and management of multistep, asynchronous, multicomponent workflows.

o   Software discovery services discover and select the best software implementation and execution platform based on problem parameters.

o   Collaboratory services support the coordinated exchange of information within potentially large user communities, either synchronously or asynchronously. They must also address security, policy, and accounting issues

o   Community authorization servers enforce community policies governing resource access, generating capabilities that community members can use to access community resources. These servers provide a global policy enforcement service by building on resource-layer information and management protocols and security protocols in the connectivity layer.

o   Community accounting and payment services gather resource usage information for the purpose of accounting, payment, and/or limiting of resource usage by community members.

 

§  Collective functions can be implemented as standalone services or as libraries designed to be linked with applications. In both cases, their implementation can build on resource layer (or other collective layer) protocols and APIs.

 

Application Layer

 

§  Application layer defines protocols and services that are targeted towards a specific application domain or class of applications.

§  Comprises the user applications that operate within a VO environment.

§  Applications are constructed in terms of, and by calling upon, services defined at any layer.

§  At each layer, we have well-defined protocols and APIs that provide access to some useful service:

o   resource management,

o   data access,

o   resource discovery, etc.

§  Applications may in practice call upon sophisticated frameworks and libraries.

§  These frameworks may themselves define protocols, services, and/or APIs.