Building an Intelligent IT Infrastructure - InformationWeek

InformationWeek is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

IoT
IoT
Government // Enterprise Architecture

Building an Intelligent IT Infrastructure

To address the complexity of today's requirements, IT needs a fresh approach to infrastructure. New technologies are coming together to help you construct a smarter, more flexible environment. Here's how to progress toward your goal.

The mounting complexity of today's IT infrastructure and the draining effect it's having on IT resources is well documented. In an effort to overcome this complexity crisis, organizations are looking to a new class of solutions commonly referred to as intelligent infrastructure. These solutions offer the hope of allowing enterprises to more easily develop, deploy, and manage their business-critical applications.

What's an intelligent infrastructure? Is it another technology to be layered into today's environment, or something beyond infrastructure as we know it today? What are the barriers to adoption, and are there any strategies and methodologies to overcome them?

The best approach to unraveling these questions is to review what organizations hope to achieve in the pursuit of intelligent infrastructure, as well as the technology options available to manage IT complexity. It's also important to build an effective plan of attack to successfully introduce these technologies and realize their benefits for your organization.

Ingredients of Intelligence

Interest in intelligent infrastructure is based on the desire to display characteristics that are difficult to achieve simultaneously through traditional IT architectures. Organizations want an infrastructure that is:

Dependable. First and foremost, the notion of intelligent infrastructure promises to yield more dependable applications. Used here, dependability is a composite of the more familiar notions of availability, reliability, and scalability. In other words, these applications must not only be available to receive new work: They must also successfully complete the work they've accepted and do this across a wide range of operating requirements. In the new world, the applications would inherit their dependability from the intelligent infrastructure — reducing the burden currently placed on developers to code their applications specifically for dependability.

Manageable. Intelligent infrastructure should allow organizations to escape the cycle of complexity that has traditionally trapped them. With intelligent infrastructure, the layers of hardware and software required to deliver dependable applications would be dramatically reduced. The intelligent infrastructure would be self-managing and would automatically respond to events, such as hardware failures.

Adaptable. Under the intelligent infrastructure paradigm, applications would be loosely coupled to the infrastructure on which they run. The infrastructure would be analogous to a utility that the application uses to do its work, as opposed to a rigid structure that has been erected to house it. When new application functionality is developed, it would be simply and easily deployed to the same intelligent infrastructure. When more application capacity is needed, organizations could add more of this infrastructure.

Affordable. In today's IT reality, the notion of dependability, manageability, and adaptability are of little significance without affordability. A truly intelligent infrastructure wouldn't rely on expensive redundancy to ensure these characteristics. Instead, it would ensure that these characteristics hold true even with inexpensive, standards-based components.

Fortunately, these objectives are becoming achievable. A variety of vendors, including major IT suppliers and service providers are offering a variety of solutions that can deliver on one or more of these objectives.

First to Second Generation

There's an irony associated with the ongoing battle against complexity. As organizations work to meet today's service-level commitments, reduce today's costs, and manage today's complexity, they often implement solutions that increase overall architectural complexity at the expense of tomorrow's costs and manageability.

We can identify several technologies that organizations are deploying to build an intelligent infrastructure. We will come to see the following technologies as first-generation solutions:

Automation and configuration management. To those steeped in the complexities of today's computing architectures, a well-understood and comprehensive set of scripts automating common infrastructure management tasks is like an oasis in the desert. Vendors have taken this notion a step further to offer a good number of deployment or configuration management products. The complexity inherent in application server products has gone a long way to boost the popularity of configuration management products, which extend the benefits of scripting while offering simplified interfaces and built-in support for multiple infrastructure elements. As these solutions become more sophisticated, vendors will build in support for interoperability standards such as the Data Center Markup Language (DCML).

Autonomic computing. Autonomic computing has gained ground as a potential solution to complexity problems through its ability to produce systems that exhibit biological characteristics, such as the ability to self-diagnose and correct problems. IBM has been perhaps the greatest proponent of this approach, but Hewlett-Packard and other systems vendors are also active. IBM's Autonomic Toolkit contains a number of tools that add intelligence to common processes, such as software installation and log file analysis.

Grid computing. The precise definition of grid computing is a moving target. However, the central concept — using decentralized, heterogeneous resources to solve computationally intense problems — has found increasing popularity, particularly in research and the sciences. These solutions emphasize the distributed nature of the computing resource, often by utilizing unused CPU cycles from nondedicated computers. Grid, therefore, helps organizations adapt quickly to changing demands on their computing infrastructures.

Virtualization and rapid provisioning. These solutions come in a variety of flavors, each with the goal of decoupling applications from the servers on which they run. Virtualization software allows companies to run multiple applications (on multiple operating systems) on a single server or mainframe. It's often used by organizations that prefer a scale-up approach. Rapid provisioning software takes a different approach. This software allows the operating system and applications, on a single machine, to be rapidly swapped out. It's often used with commodity class computers, such as blade servers, as part of a scale-out approach.

Table 1 offers a comparison of first- and second-generation technology approaches. While each solution allows organizations to reap efficiencies over traditional approaches, first-generation solutions are best suited to alleviating specific organizational pain points. In our opinion, they fail to address the root issue of complexity at its core, which limits them to providing incremental benefits at the expense of future manageability and cost.

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.
Previous
1 of 2
Next
Comment  | 
Print  | 
More Insights
Slideshows
7 Technologies You Need to Know for Artificial Intelligence
Jessica Davis, Senior Editor, Enterprise Apps,  7/1/2019
Commentary
A Practical Guide to DevOps: It's Not that Scary
Cathleen Gagne, Managing Editor, InformationWeek,  7/5/2019
Commentary
Diversity in IT: The Business and Moral Reasons
James M. Connolly, Editorial Director, InformationWeek and Network Computing,  6/20/2019
White Papers
Register for InformationWeek Newsletters
Video
Current Issue
Data Science and AI in the Fast Lane
This IT Trend Report will help you gain insight into how quickly and dramatically data science is influencing how enterprises are managed and where they will derive business success. Read the report today!
Slideshows
Flash Poll