Autonomic Computers

	Log In

Search Options

Autonomic Computers

Introduction

What Is Autonomic Computing?

Taking it to the Bank

Vendors Offering Autonomy

The Semantics of Autonomy

Introduction

When I encounter a scary situation, my heart begins to race and adrenaline pumps into my system to give me the energy I need to react to the scary stimulus. These bodily responses are initiated automatically by my autonomic nervous system—a self-regulating system of organs and muscles that controls our sleep, digestion, and fight-or-flight responses.

When a computer system needs to react to a stimulus, such as a hacker, an upgrade, or a change in business requirements, what are your options? Does a human operator have to intervene, or can the system develop autonomic capabilities similar to our nervous system? Several IT vendors already offer tools that help computers become more self-sufficient and less reliant on the workers that have to maintain them. However, the ultimate autonomic dream—a system that can modify itself for changes in business strategy—is still far, far away.

A wide variety of applications can benefit from these self-managed systems. Peter Hughes, IT Division assistant chief for technology at Maryland-based NASA's Goddard Space Flight Center, calls autonomic computing a "breakthrough technology." Its use in deep space missions is promising for the Center. However, Hughes adds that there are many challenges to overcome in several areas, such as scalability and handling compound problems that affect numerous functions.

What Is Autonomic Computing?

Autonomic computing already frees up IT workers from mundane, repetitive tasks, such as restoring a failed system or moving around network resources. The focus for the next few years for vendors will be to deliver systems and networks that feature self healing capabilities across resources and applications. According to experts, a computing infrastructure that modifies itself for business strategy and operational needs is likely a decade away.

As a market/research leader, IBM sees autonomic computing in five levels or stages. Stage One contains no coordinated systems management. Stage Two offers management tools that are centralized and require human intervention. Systems begin to analyze problems and suggest solutions in Level Three. In Level Four, systems correct problems without humans. Level Five brings in an organization's business objectives so that the system can manage to those objectives.

Taking it to the Bank

The reality for CPAs and accounting professionals who provide consulting services is to offer their clients, employers, and customer's specific information about how autonomic computers can help an organization save time, money, and other resources. Because of the newness associated with this technology, some of the long-lasting effects are still not completely known. However, experts already are projecting that the potential for cost savings from using autonomic computers is enormous.

A Forrester Research study, for example, reports that server utilization rates could jump from the current average of 20 percent to 80 percent or higher. With a server capable of constantly monitoring its own efficiency instead of a network analyst who would be bored performing this task, companies will be able to do much more with much less computing power, as well as less labor. In addition to reducing labor costs, this technology is predicted to significantly speed up response times and overall system efficiency by using less computer power more productively.

Vendors Offering Autonomy

Already, there are a number of well-known vendors that have developed solutions.

· IBM has developed autonomic system management tools to help with data center management and system optimization. A beta version of its DB2 Universal database was released in May 2004. This product contains autonomic database performance optimization, a tool that enables databases to respond to user information requests faster. It also includes other methods to make the database more efficient for the users: a client rerouting capability to reduce downtime and learning optimization technology where the software actually learns and improves its system corrections. All of these autonomic features are designed to improve data access times for the database user. IBM offers many more self-managing products, and in 2004, it offered seminars throughout the country on autonomic computing.

· HP offers dozens of IT management tools that it terms as "autonomic." Last year, it announced the new Systems Insight Manager, a systems management software capability that works on Microsoft (MS) Windows, Linux, and Unix systems. Its Virtual Server Environment allows managers to plug in service level objectives, such as uptime percentages that the system must meet by autonomically shifting resources. For example, in the case of an uptime requirement, the system can replace a server that has gone down to keep the system running smoothly.

· ZENworks is Novell's product for IT management automation, offering resource management for desktops, servers, laptops, and handheld devices within the organization. ZENworks supports multiple platforms, including MS Windows, Linux, Solaris, and Netware.

· Computer Associates International, Inc. has offered Unicenter Enterprise Management for some time now. This system automates day-to-day IT management tasks, proactively identifies problems, automates software installation, and dynamically allocates resources, such as replacing a failed hard drive or adding more servers when a spike in demand is detected.

· Cisco offers its Self-Defending Network strategy to autonomically attack network security threats.

The Semantics of Autonomy

Autonomic computing is beginning to infiltrate our technology environment in other ways as well. For example, numerous vendors are including the word "autonomic" in their sales literature. A sampling of vendors that are working on autonomic initiatives include Sun Microsystems, Inc., BMC Software, Inc., Microsoft Corp., and EMC Corp.

In relation to autonomic computing, you might see the phrase "self-healing" in computer literature. Some analysts say self-technologies comprise autonomic computing, but actually also include resource and utility management, while autonomic computing focuses on problem identification and resolution. In fact, the definition of autonomic computing given in the AICPA Top 10 Technologies program includes the term "self-healing:"

"[Autonomic computing encompasses] tools and strategies to manage and maintain all systems across the enterprise, including system maintenance, upgrades, automatic patching, and self-healing (for example, ZenWorks, Unicenter management alerts). This is an approach toward self-managed computing systems with a minimum of human interference. The term derives from the body's autonomic nervous system, which controls key functions without conscious awareness or involvement."

IBM considers an autonomic system to be able to perform four major functions: Self-healing, Self-optimization, Self-configuring Self-protecting.

Another critical aspect of autonomic computing is to meet certain business policies and rules. For example, the system must be able to manage itself through a set of policies and objectives and know how to react to input controls, including service levels, resource objectives, user priorities, and other business policies.

Some definitions of autonomic computing include the function grid computing, which is a cluster of networks, computers, and databases that work together to form a supercomputer. Control of the grid infrastructure is centralized and automated. Although the lines are blurry between the terms "autonomic" "self-healing," we can arrive at a generally accepted agreement about what autonomic computing is: the concept of a systems intelligence that runs an enterprise-wide systems infrastructure with limited human intervention. It also is run on predefined business policies and objectives.

One day, CPAs will be able to describe the business processes they want automated, and the systems will simply assemble themselves. That's the vision of IBM's Vice President of Autonomic Computing, Alan Ganek, and the potential of autonomic computing in our future. For now, CPAs can be content with potential savings in network hardware and labor, and the knowledge that on the horizon, their business's computing infrastructure will become increasingly easier to manage.