DAILY NEWS AND INFORMATION FOR THE GLOBAL GRID COMMUNITY / JUNE 23, 2003: VOL. 2 NO. 25    ( Table of Contents )

Special Features:

GRIDS - A VERTICAL MARKET PERSPECTIVE PT. 1
Insight Research

Grid computing is a form of distributed system that makes it possible for computing resources to be shared across networks. Just as Web standards and technologies enabled universal, transparent access to documents, Grid promises to do much the same for computing resources. Grid enables the selection, aggregation, and sharing of information resources resident in multiple administrative domains and across geographic areas. These information resources can be shared based upon their availability, capability, and cost, as well as the user's quality of service (QoS) requirements. The advantages of a grid architecture include:

• reducing the total cost of ownership (TCO);
• aggregating and improving the efficiency of computing, data, and storage resources; and
• enabling the creation of virtual organizations for applications and data sharing.

In the US alone, several hundred million dollars of government funding were spent on grid research initiatives since the mid-1990s. Elsewhere in the world, there are currently about 20 to 30 substantial academic/research grids under development or in production. The accent today, however, has shifted to commercialization of grid technologies for the enterprise.

Grid computing is but one of several fundamental technologies that many analysts predict will transform the IT industry and the enterprise IT infrastructure over the next decade. As enterprises of all kinds are demanding better IT resource utilization, lower administrative and management costs, and distributed, failure-tolerant infrastructures, a grand synthesis of these technologies is in progress. The resulting IT infrastructure is expected to be radically different from today's, with lower costs, failure- tolerant systems, and highly distributed, inter-enterprise applications.

As of yet, there does not appear to be a commonly agreed upon name for this synthesis. Prior to their acquisition by IBM, PricewaterhouseCoopers Consulting called the service grid "Internet Computing," IBM now calls it "e- business on demand," and yet others call it "Next-Generation Data Center" or "P2P Computing."

What's Driving Grid Computing?

Since grid computing uses open protocols and standards to perform distributed computing over the Internet or private Intranets, it is clear that grids focus on resource sharing and virtualization. Interest in grid technology is being driven by a number of key concerns of IT customers across all industries, geographies, and sizes of business, including:

Increased computational needs Enterprises, beyond the traditional scientific community, are increasingly running compute-intensive and high-throughput applications. These applications generate dramatic spikes and lulls in demand for computing power, and the IT department is challenged to engineer the balance of appropriate capacity level and cost.

Data chaos Huge amounts of dispersed data and data silos, as well as multiple users who need to share data, create data management challenges. Constant data updates cause version-control nightmares, particularly where the data and users are widely dispersed.

Distributed and virtual organizations Collaboration is a key driver of grid technology adoption across many vertical markets. The need for widely dispersed work teams to work closely together is increasing within enterprises, often as a result of mergers or acquisitions, but also due to frequent internal restructuring. Additionally, linkages among enterprises continue to multiply rapidly, particularly in supply chains and e-commerce exchanges. The business need for collaboration, in turn, drives the technical need to effectively utilize geographically distributed resources.

Underutilized IT resources As enterprises struggle to cut costs, one area that stands out is underutilized IT resources. According to a number of industry studies, even in most the efficient data center, 70 to 80 percent of processing power often sits idle. Although mainframe technology has evolved over three decades to insure high levels of utilization and efficient use of resources, the situation for servers and desktops is very different. As a result, many enterprises are beginning to experiment with Grid technologies given such underutilized resources. Applications can be run on any available and appropriate server through deployment of a grid, and vendors claim that utilization rates can surpass 90 percent.

Despite the many benefits that are possible, Grid is in its infancy today, and is not without drawbacks. Very few applications have been enabled to take full advantage of the Grid. Security issues are paramount; the reality on the ground today is that grids are being used first within single organizations, and this is likely to be the trend for the next several years.

Using Grids as a Tool for Resource Sharing

Grids can be categorized by the type of resource that is shared, and grid projects, systems, and applications can fall into one or more categories.

Compute Grids The original concept for grid technology arose from the need to provide high-performance computing on demand to the academic and research communities. Grid infrastructures were developed to deliver supercomputing power to large scientific projects with complex number-crunching applications that required a vast number of calculations. Compute grids provide high- throughput computing through the coordinated use of many computers that are usually geographically distributed.

In a compute grid, a single application is split into smaller pieces to run on many different computers simultaneously, producing supercomputer speed from off-the-shelf hardware. Compute grids can significantly improve the speed and efficiency of executing applications that involve complex and compute- intensive modeling, simulations, and animations.

Compute grids enable an organization to harvest spare cycles on servers, workstations, or desktops, or on some combination of these. In many enterprise grid projects that are initiated by the IT department, the interest is primarily in making optimal use of server cycle rather than PC cycles, since the server environment is generally more controlled and the IT department perceives less business risk. To date, the vast majority of grid computing research and commercial product development has focused on compute grids.

Data Grids Data grids involve accessing geographically distributed data. The term "information grid" is sometimes used synonymously with data grid. Data grids take grid computing to the next level, beyond a means of increasing computing power to a means of collaborating and sharing data and information resources. As a result, data grids can facilitate collaboration while protecting valuable intellectual property.

The need to share massive amounts of data as well as computing resources across many locations is typical of many academic and research efforts. Enterprises increasingly need to share large files and data sets across multiple locations as well. For example, a typical pharmaceutical company has research teams around the world that must share data. Most enterprises solve the problem of sharing data access across a wide area through the use of manual processes and tools such as file transfer protocol (FTP). Data grids offer the promise of providing simplified data access across multiple locations and systems, by providing distributed management of large quantities of data.

Instrumentation Grids Instrumentation grids provide shared access to expensive and/or unique scientific instruments such as radio telescopes and electron microscopes for near real-time data processing and analysis. To date, instrumentation grids are of peculiar interest to the research and academic communities. As life sciences and manufacturing processes of all kinds become more highly automated, however, there should be increased interest within IT departments and enterprises.

Application Grids Application grids are the least well-defined and developed of the four resource classes of grids. In concept, they ensure secure, wide- area application access and utilization. In today's IT world, application grids hold significant promise of decreasing the complexity of developing and implementing cross-enterprise and multi-enterprise applications.

It is likely that once Web services become more wide-spread as a method for application development and interoperability, the resulting disaggregation of application components will set the stage for the emergence of application grids.

GRIDtoday, HPCwire, and DSstar subscribers can look forward to a discounted rate on this detailed Insight Resarch report, starting June 30th, 2003. Be sure to look for our special offer in the upcoming issue! For more information or to order this report send a detailed message to reports@tgc.com or call 858-625-0070.