Basic Concepts

From ScenarioThinking
Jump to navigation Jump to search
  • What is GRID?

In 1998, grid gurus Ian Foster and Carl Kesselman defined the grid as a hardware and software infrastructure that provides dependable, consistent, and pervasive access to high-end computational resources.
· IBM's definition of Grid is "all or some of a group of computers, servers and storage across an enterprise, virtualized as one large computing system".
· For Oracle's Larry Ellison, the grid is "capacity on demand. Plug another server into the grid and the application runs faster and more reliably."


  • What is GRID computing?

Grid computing is a form of distributed computing that involves coordinating and sharing computing, application, data, storage, or network resources across dynamic and geographically dispersed organizations.
Ian Foster and Carl Kesselman stated that grid computing is concerned with coordinated resource sharing and problem solving in dynamic, multi-institutional virtual organizations.
However,the the definition of GRID computing keeps shifting. The technologies and standards surrounding grid computing are still evolving.
More detailed definition can be found here:
http://www-fp.mcs.anl.gov/~foster/Articles/WhatIsTheGrid.pdf


  • What is GRID architecture / infrastructure?

GRID unifies all existing technologies in this area, such as clustering, peer-to-peer. It lies on top of the infrastructure by providing modules that form the Grid operating system. Major components include:

  1. a dynamic resources directory,
  2. a distributed data management system,
  3. an application runtime and monitoring system,
  4. informaiton services, and
  5. security services.

Above this infrastructure are many applications GRID technology provide. More detailed information please refer to: http://www.globus.org/research/testbeds.html


  • What is the difference between GRID & Cluster?

Cluster can constitute a powerful Grid resource. However, such a system is not a Grid itself, due to its centralized control of the hosts that it manages: it has complete knowledge of system state and user requests,and complete control over individual components.
GRID, unlike cluster, integrates and coordinates resources and users that live within different control domains—geographically distributed and heterogeneous.


  • Who use the GRID now/ in the future? ?

Currently, Butterfly.net, Inc., and IBM will deploy the first-ever custom commercial grid for the online video gaming market. The Butterfly Grid™ could enable online video game providers to support a massive number of players within the same game by allocating computing resources to the most populated areas and most popular games.
Moreover, many scientific institutions are in busy research of implementing Grid technology and finding its new applications.
Some examples of industry that can benefit from Grid in the future include: Automotive and aerospace, for collaborative design and data-intensive testing; financial services, for running long, complex scenarios and arriving at more accurate decisions; life sciences, for analyzing and decoding strings of biological and chemical information; government, for enabling seamless collaboration and agility in both civil and military departments and agencies; higher education for enabling advanced, data and compute intensive research.


  • What are the possible benefits (motives) of implementing GRID?

Benefits can be extensive. They include:
· Accelerated time to results, which allows for the provisioning of extra time and resources to solve problems that were previously unsolvable

· Improved productivity and collaboration

· Allowing widely dispersed departments and businesses to create virtual organizations to share data and resources

· More flexible, resilient operational infrastructures

· Instantaneous access to compute and data resources to "sense and respond" to needs

· Leveraging existing capital investments, which helps to ensure optimal utilization of computing capabilities

· Avoiding common pitfalls of over-provisioning and incurring excess costs

· Freeing IT organizations from the burden of administering disparate, non-integrated systems

. Enormous computation resources for everyone.


  • What other technologies are needed to help its development? the relationship between them?

GRID can be seen as the latest and most complete evolution of more familiar developments, such as: distributed computing, the Web, clustering, peer-to-peer and virtualization technologies.
· Like the Web, grid computing keeps complexity hidden: multiple users enjoy a single, unified experience.

· Unlike the Web, which mainly enables communication, grid computing enables full collaboration toward common business goals.

· Like peer-to-peer, grid computing allows users to share files.

· Unlike peer-to-peer, grid computing allows many-to-many sharing — not only files but other resources as well.

· Like clusters and distributed computing, grids bring computing resources together.

· Unlike clusters and distributed computing, which need physical proximity and operating homogeneity, grids can be geographically distributed and heterogeneous.

· Like virtualization technologies, grid computing enables the virtualization of IT resources.

· Unlike virtualization technologies, which virtualize a single system, grid computing enables the virtualization of vast and disparate IT resources.


  • What are the challenges?

The major challenges are, concerning security (authentication, authorization etc.), resource allocation, scheduling, access of remote data sets, policy management, fault-tolerance, failure recovery, site autonomy, QoS (Quality of Service). Additionally, the Grid must meet stringent, dynamically changing performance requirements and be far more dependable than current peer-to-peer networks.
There are thorny social and political challenges too. Building a seamless worldwide infrastructure requires global consensus on grid protocols—the basic mechanisms by which users negotiate, establish, and exploit shared access to resources.


  • Grid Computing Vs
    • Cluster Computing: Cluster computing is about resources aggregation in a single administrative domain whereas Grid computing in about resource sharing and aggregation across multiple domains.
    • P2P Vs Grid: It's different from peer-to-peer in respect that it must provide the computational resources in a dependable way. Also, P2P does not provide the level of security features, authentication and virtual machine architecture that grid will.
    • Distributed/Internet Computing: Distributed Computing projects like Seti@Home or Genome@Home are not in true sense grid computing because they are neither real-time nor online.
    • Grid is not just Powerful Internet: A grid is not simply a more powerful Internet, or Internet-2. It has some unique characteristics like dynamic sharing (DS), single authentication (SA), and virtual collaborations (VC).

Dynamic sharing is not P2P or client-server. DS is an ability to add or change sharing during operation of an application, either in response to changing requirements, or from a participant's command. Single authentication means that you identify yourself once, and that authorization is automatically forwarded where needed.


  • Others Perspective on Grids
    • Grid: As An Utility

Grid is sometimes envisioned as an utility model over the Internet, where clients draw on compute power much as they do now with electricity. The systems would work much like electrical utilities, giving companies access to virtually unlimited surplus computing power, regardless of where it's stored, and make it easier to share stored data and specialized software with partners on the grid.

    • Grid: The Next Generation Internet

The grid is not an alternative to internet but rather a set of additional protocols and services build over internet to support collaborative-computations

    • The Grid makes high-performance computers superfluous?

If we have the chunk of computing powers available from Virtual Organizations, isn’t it true that it will make high-performance computers out of business? Well, not really as 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.




>>back>>