Grid is a type of parallel and distributed system that enables the sharing, selection, and aggregation of resources distributed across "multiple" administrative domains based on their (resources) availability, capability, performance, cost, and users' quality-of-service requirements.
The vision is that the Grid will be to computational resources what the World Wide Web is to documents containing information.
The drivers for the Grid have come from the scientific community, who need increasing computational power, e.g. Human Genome Project. It is infeasible to assemble the computational resource needed for large scale analysis at a single location.
The one trait that the new scientific and business computing environments share is the need to manage dynamic, distributed infrastructures, services, and applications.
The main industries needing Grids are:
These industries have particularly high requirements for computational power in their research departments.
Resource aggregation - allow corporate users to treat a company’s entire IT infrastructure as one computer, grabbing unused resources as they are needed
Data sharing - allow companies to access remote data. TThis is of particular interest in certain life sciences projects in which companies need to share human genome data with other companies
Collaboration via Grids - allow widely dispersed organizations to work together on projects, integrating business processes and sharing everything from engineering blueprints to software applications.
The real and specific problem that underlies the Grid concept is flexible, secure coordinated resource sharing among dynamic collections of individuals, institutions, and resources. These collections are defined as virtual organizations (VOs), and this concept is fundamental to the problem that Grid technologies attempt to address.
The sharing that we are concerned with is not primarily file exchange but rather direct access to computers, software, data, and other resources, as is required by a range of:
In business, concrete examples of companies that are already involved in this way with enterprises are application service providers (ASP) and storage service providers (SSP)
The requirements for virtual organisations are:
Effective VO operation requires that sharing relationships can be established among any participants, inferring that interoperability is the central issue for Grid technologies - interoperable across organisational boundaries, operational policies, and resource types. Hence, the Grid architecture is first and foremost a protocol architecture, with protocols being used to define the mechanisms by which VO users and resources negotiate, manage, and exploit sharing relationships.
In specifying the various layers of the Grid architecture, the principles of the “hourglass model” are used. The narrow neck of the hourglass defines a small set of core abstractions and protocols (e.g., TCP and HTTP in the Internet), onto which many different high-level behaviours can be mapped (the top of the hourglass), and which themselves can be mapped onto many different underlying technologies (the base of the hourglass).
The Grid Fabric layer provides the resources to which shared access is mediated by Grid protocols: for example, computational resources, storage systems, catalogues, and network resources.
The Connectivity layer defines core communication and authentication protocols required for Grid-specific network transactions.
The Resource layer builds on Connectivity layer communication and authentication protocols to define protocols (and APIs and SDKs) for the secure negotiation, initiation, monitoring, control, accounting, and payment of sharing operations on individual resources.
While the Resource layer is focused on interactions with a single resource, the next layer in the architecture contains protocols and services (and APIs and SDKs) that are not associated with any one specific resource but rather are global in nature and capture interactions across collections of resources. For this reason, we refer to the next layer of the architecture as the Collective layer.
The final layer in our Grid architecture 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.
Current distributed computing technologies do not accommodate the range of resources types or do not provide the flexibility and control on sharing relationships needed to establish VOs because:
Grid technologies complement rather than compete with existing distributed computing technologies
The Global Grid Forum (GGF) announced the Open Grid Services Architecture (OGSA), a set of specifications and standards for creating “Grid services” (defined below) using the Globus toolkit protocols and the web services standards of XML, SOAP and WSDL. This has been primarily driven by IBM and Globus.
In OGSA, computational resources, storage resources, networks, programs, databases, and other resources are all represented as services. It builds upon the concepts and technologies of the Grid and Web Services communities to provide an architecture that defines a service semantic – Grid services. OGSA:
Companies such as Microsoft, IBM, and many of the startups (e.g. Avaki, Entropia) support OGSA
More information at http://www.globus.org/ogsa/
The Grid concepts provide a clearly separate layer of protocols that address operational integration, in support of the web services protocols which address interaction (application integration). The Grid architecture should be viewed as a key enabler of service based architectures.
Some key problems the Grid can address:
· Enabling B2B communications
· Ensuring Quality of Service (QoS)
· More effective outsourcing of computing power
However, there are still many obstacles which have not been resolved yet, e.g. security – if my job is running on someone else’s systems, is my data safe ?
Grid technologies have the potential to achieve significant savings through more efficient use of hardware and ease of collaboration between enterprises. There is also a complementary opportunity to grow revenues through being able to work effectively in virtual organisations. However, the grid technologies are still immature.
Very few enterprises have deployed grids, and today data cannot be used on a shared (i.e. inter-enterprise) grid without a loss of control. The only opportunity today is for enterprises to deploy grids internally to manage their hardware better, but this would only be cost-effective if the enterprise had significant computational requirements, i.e. research.
Globus Toolkit (www.globus.org)
The Globus Toolkit is a community-based, open-architecture, open-source set of services and software that support Grids and Grid applications. It is the de facto standard for Grid environments. There is a significant amount of grid technology information on this site, including the latest research.
Avaki (www.avaki.com)
Data Synapse (www.datasynapse.com)
Entropia (www.entropia.com)
GridFrastructure (www.gridfrastructure.com)
Parabon (www.parabon.com)
Platform Computing (www.platform.com)
United Devices (www.ud.com)
IBM (www-1.ibm.com/grid/)
Sun (wwws.sun.com/software/grid)
HP (http://devresource.hp.com/topics/utility_comp.html)
Oracle (http://otn.oracle.com/products/oracle9i/grid_computing/content.html)
IBM is acknowledged as being the leading major software vendor in grid computing, in terms of its financial investment, use of open standards, and vision. Microsoft does not have a Grid offering yet.
Global Grid Forum (www.gridforum.org) is leading the Grid community in the promotion and development of Grid technologies and applications, development of best practices, and establishing some loose standards (i.e. not rigidly specified standards)
Grid Computing Planet (www.gridcomputingplanet.com) - news site
Grid Today (www.gridtoday.com/) - news site
Grid Computing Info Centre (www.gridcomputing.com) - list of current Grid projects