Home > Articles > Operating Systems, Server > Solaris

  • Print
  • + Share This
Like this article? We recommend

Overview of a Global Compute Grid

In this section, these basic questions are briefly addressed.

  • What are the motivations for creating a grid?

  • What are the major areas of difficulty that must be addressed?

  • What are the currently available components of a global compute grid and how do they solve the issues?

The Globus project introduces the term virtual organization (VO) as a set of users in multiple network domains who wish to share some of their resources. The virtual organization might be (as is currently the norm) a group of academic institutions that wish to enhance resource sharing. The possible functional objectives of such resource sharing include:

The aggregation of compute power can bring a number of benefits:

  • Increases the throughput of users' jobs by maximizing resource utilization

  • Increases the range of complementary hardware available, for example, compute clusters, large shared memory servers

  • Provides a grid supercomputer that can provide a platform for grand challenge applications

  • Provides the tight integration of geographically and functionally disparate databases

  • Provides the catering for a huge, dynamic data set and the processing there of.

The main pieces of a global compute grid are:

  • Interface—This enables access to the grid by nonexpert users. Some straightforward interface usually based upon a web-portal must be created.

  • Broker—This is automating job scheduling based upon the user's policies. Such policies could describe the user's priorities in terms of job requirements, available budget, time requirements and so on. The broker would use these policies when negotiating on the user's behalf for a resource on the grid.

  • Security, data-management, job-management, and resource discovery—These are the key issues that have been addressed by the Globus project. They are described further in the discussion of the Globus Toolkit.

  • Resource guarantees and accounting—These are areas of current intensive research activity, and links with the brokering technologies. In this BluePrints article, the features of the DRM are used to help solve these issues at the cluster grid level.

In this article we describe the components of a simple compute grid where geographically dispersed compute and storage resources are brought together and presented to users as a unified resource. First, the general concepts are discussed and then two specific implementations are described. The middleware discussed and implemented in the two case studies is the Globus Toolkit version 2.0 or 2.2.

  • + Share This
  • 🔖 Save To Your Account