First generation of GRID :

FAFNER (Factoring via Network enabled recursion):  This is a set up to factorize RSA130 using Number Field Sieve (NFS) using computational resources of web-servers.

NFS is implemented in such a way that it does not require any communication between nodes, and is such efficient that it can be run in a workstation with 4Mb of memory. CGI scripts on server side are invoked by the contributors.

I-WAY: This is a project to integrate High-performance Computing device with existing High bandwidth networks. It uses ATM(Asynchronous Transfer mode), supporting both TCP and direct ATM. It uses I-POP servers acting as a gateway to I-WAY. I-POP servers are UNIX workstations possessing standard software environment.

Each site participating in I-WAY ran I-POP server. The I-POP server allowed uniform access, authentication, resource reservation, process creation and communication functions.

N.B. AFS: Andrew File system, a distributive file system that enables co-operating hosts to share resources in LAN or WAN.

Second Generation of GRID :

FAFNER was forerunner of SETI@home and Distributed.NET

I-WAY for Globus and Legion

Second Generation grids have issue like Heterogeneity, scalability, Adaptability.

Globus is a software infrastructure or can be said as a layered architecture in which global services are built upon the core-local services.

  • For resource allocation and monitoring HTTP based GRAM(Globus Resource Allocation manager) is used.
  • For High speed transfer Extended FTP protocol, GridFTP is used.
  • Authentication GSI (Grid Security infrastructure)
  • LDAP is for distributed access to state and information. (Lightweight Directory Access Protocol, build on TCP for querying state of remote Database)

GRID Portals :

A GRID portal will provide access to certain kind of resources, For example a grid portal will authenticate users, permit them to access remote resources, help them to schedule jobs, and allowed to manipulate the resources information.

Scalability Issue:

Client server model can be a performance bottleneck and a single point of failure, so a plausible approach is to resolve the issue with decentralization, however it does not seems to be simple solution. P2P computing machines share data, resources via internet.

Some obstacles in P2P are communication and security overhead, security exploitation. Sometimes finding huge number of resources from globally based decentralized peers is difficult.

THIRD Generation:

This is a service oriented architecture, it standardizes

  • Web services (SOAP, XML protocol, WSDL(Web services description language), UDDI(Universal description discovery and integration)).
  • OGSA framework

So, third generation system service oriented approach uses metadata feature and can also show autonomic features.


I-WAY high performance distributed computing experiment

I-WAY focuses on interactive environments and backend super-computing to couple users more tightly with computers. I-WAY has been developed keeping in mind execution of large application; applications using virtual environments with remote supercomputers, data systems, or scientific instruments.

Geographically distributed supercomputers which are highly heterogeneous, and the networks is also heterogeneous. Most networks uses ATM, whereas CASA uses HIPPI technology. I-WAY standardized TCP/IP for application networking.

I-WAY provides solutions to the following requirements:

  • Resources naming ,
  • Uniform programming environment
  • Autoconfiguration and resource characterization dynamically
  • Distributed data services
  • confidentiality and integrity

I-POP machines provide a uniform environment (I-Soft) for deployment of management software and also security, scheduling, parallel programming support. I-POP’s ‘Autheticate-once’ capability is extremely useful. It uses Kerberos method of authentication.

I-WAY uses Filesystem which provide many users to access various data and utility programs at many different sites(multiple user). Second, users running programs on remote computers must be able to access executables and configuration data at many different sites(higher performance). Third, application programs must be able to read and write potentially large data sets(all above requirements).


Implementing Production Grid:
Grid software is not a monolithic package but collection of interoperating software packages or modules.

The Anatomy of GRID
The real and specific problem that underlies the Grid concept is coordinated resource sharing(direct access to computers, data, computing and other resources) and in dynamic, multi-institutional(sharing rules must be defined) virtual organizations (VO). It is obvious that one Physical organization can participate in multiple Virtual Organizations(VOs). Interoperability is a major concern for GRID Architecture, without this, the VO applications
& participants will enter in an agreement but there will be no assurance that the mechanism used by any other parties are also same. So we require standard protocols, syntaxes, and ‘middleware’ provides this.

A glimpse on relationship with Grid Protocol and Internet Protocol

Fabric layer : Interfaces to local control, it provides mechanism that manages

  • Computational resources, monitoring and controlling processes.
  • Storage resources
  • Network resources
  • Code repositories
  • Catalogs

Connectivity: Transport, routing and naming.

Resource : This layers defines protocols for secure negotiation, initiation and monitoring, control, accounting, sharing of individual resources.

Collective: Coordinating multiple resources, like Directory services, scheduling, monitoring services, workload management etc.

Collective and resource layer protocols, services, APIs, and SDKs can be combined in a variety of ways to deliver functionality to applications

Relationship of GRID with other technologies :

World wide web :

If single sign on capabilities is provided by GSI extension to TLS integrated in Web-browsers then it will provide single sign-on to multiple web servers.

Enterprise Computing System:

CORBA, Java Beans, J2EE, are all system designed to build distributive applications.The sharing arrangements in these applications are really static, Grid can provide CORBA an Object Resource Broker(ORB) that uses GSI to address Cross Organizational Security issues, and Portable object Adaptor will manage the resource across the VOs.

Some perspective with GRIDs:

  • Next gen Internet, Set of additional services and protocols built on Internet protocols will enrich the computational and data environment on the Net.
  • Grid means controlled sharing, Resource owners can enforce policies whether to share resources with a group of members or not, and can also log accounting information.
  • Distributive Operating System
  • Abstraction and encapsulation can reduce complexity and improve reliability and that is the type of programming models Grid requires. Grid resources should be presented to users as Objects, Object –oriented API’s will be best for that.
  • The Grid makes high-performance computers superfluous:  The hundreds, thousands, or even  millions  of  processors  that  may  be  accessible  within  a  VO  represent  a  significant source of computational power, if they can be harnessed in a useful fashion. This does not imply, however, that traditional high-performance computers are obsolete. Many problems require tightly coupled computers, with low latencies and high communication bandwidths; Grid computing may well increase, rather than reduce, demand for such systems by making access easier.





  1. No comments yet.
  1. No trackbacks yet.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: