In-Depth
5 Best Practices for Reducing IT Complexity
Can your business users trust your IT department? A new approach to enterprise automation may help IT better manage IT complexity and enhance that user/IT trust.
By John Connor, Chief Technology Officer, ASG
Would you trust a banker who cannot balance a bank statement or an auto mechanic whose car breaks down in front of you? The same prudence applies to an information technology (IT) organization. The business needs to be confident that the IT organization is expertly supporting its objectives. A new approach to enterprise automation -- called enterprise automation for the 21st century -- can help IT better manage its complex environment and help the business achieve its results.
Managing the complexity of an IT organization is not easy. A typical operational IT environment with multiple users across an enterprise consists of more than applications. It includes many platforms with their own set of tools, management requirements, and frequently disconnected application components (including Web services and cloud computing).
At its current pace, IT complexity is doubling every two years. We have seen dramatic increases in processing power, communications bandwidth, and data storage. For example, the largest data warehouses now are 100 terabytes (five times the amount of all text in the Library of Congress) and the fastest long-distance network links exceed 40 gigabits per second, fast enough to transmit the entire text of the Library of Congress in just over one hour.
However, this technical capacity is meaningless unless it is used to meet human and business requirements. IT managers have turned to enterprise automation to manage IT complexity. Current approaches include networks, consoles, complex event processing (CEP), and performance optimization, among others. Although each approach solves issues within its respective domains, each operates in silos and independently from each other.
A new concept called enterprise automation for the 21st century can help IT organizations deal with constant change and manage complexity. The concept suggests using information aggregation at the business-service level to link silos and better utilize technical capacities and team resources across the entire organization.
The new enterprise automation approach uses an abstraction layer coupled with a visualization layer to better manage and align IT and business functions. An abstraction layer is simply the "interface" between the user and the specifications. For example, Windows is one large abstraction layer. Dragging a file from one folder to the other is a complex operation. However, thanks to the abstraction layer, users don't have to know what operations are needed to move the file on their desktop. The operations are happening "under the covers" of the layer. Abstraction layers allow applications to be created without regard to the underlying requirements.
A New Approach
In the new enterprise automation approach, the IT infrastructure is considered a data source. The abstraction layer removes the requirement to understand or interpret the underlying data or data sources. All components that provide data must be included in the management structure, including formal tools, applications, logs, and disconnected and informal sources.
By using this approach, IT organizations can look at all the systems in the enterprise that create data and form the services that the business depends on to function. With 21st-century enterprise automation, predictive capabilities can automatically define the correct remedial action before a technical problem disrupts the business service. The system issues an alert when a business service is outside of set parameters. The business can perform a root-cause analysis to determine what is happening, why it is happening, and what else is affected to devise a solution. This new level of real-time monitoring and automation helps to reduce the time and costs previously required to locate and implement repairs to faults -- known as mean time to repair (MTTR).
Moreover, IT staff with silo expertise can visualize the impact of their respective areas on the entire enterprise, especially on business services, applications, and translations. Linking formerly isolated areas of expertise is the key to building high-performing IT teams and high-performing organizations. Employees become skilled in managing functions that are necessary to provide the highest level of customer service.
Automation Challenges
There are two primary challenges that may hinder a business from achieving the full value of enterprise automation.
First, IT has become overspecialized. Although in some cases silos are necessary, having to deal with multiple isolated silos of IT expertise makes the organization prone to enterprise-level errors and creates dissatisfaction among its consumers. The silos need a way to communicate and collaborate effectively.
Second, adapting this new approach to enterprise automation requires a change of behavior. It involves nurturing collaborations across teams, departments, and the entire organization. The previous lack of defined processes and standards forced a reliance on "heroes" -- people within the IT organization who frequently put out unexpected fires on their own. Now the business has a better approach that allows the team to work together and focus on strategic tasks proactively instead of being tied up with reactive tasks.
Five Best Practices
The most frequent question that came up in my conversations with thousands of IT users and managers over the past years was how companies can start applying this new approach to enterprise automation.
These five best practices will help you select a solution for managing IT complexity and simplify processes.
Best Practice #1: Focus on actionable information
Many tools or solutions are passive data providers. You need to make this information actionable by interpreting what the data means and what the impacts may be. Based on this assessment, you can choose the corrective action that may be required. The new solution needs to do the work and provide the direction or answer to the detected issues.
Best Practice #2: Leverage existing tools and management platform
Any new solution needs to compliment or enhance existing tools or management platforms. It should use them as data sources. Avoid costly replacements and replace tools and platforms transparently over time instead.
Best Practice #3: State a clear return on investment (ROI)
A clear value -- or ROI value -- must be realized for the IT investment. This value must take into consideration the time and resources required. The value must exceed the alternative (that is, preserving the status quo). In fact, doing nothing at all often has costs associated with it.
Best Practice #4: Develop a road map
Tied to the ROI value is the development of a road map based on a realistic understanding of the required staff, time, and financial commitment. The road map should provide an honest assessment of where the organization is today and where it wants to be in the future.
Best Practice #5: Eliminate repetitive tasks
Every solution should increase productivity by reducing human intervention and interpretation. Start by eliminating tasks that need to be done time and time again, such as restarting a process or a server after a failure. Use a solution that can do these tasks automatically and creates a performance report in real-time.
A Final Word
Following these best five best practices are the first steps towards reducing IT complexity and leveraging a new approach to enterprise automation. Today, IT needs to rely on abstraction to achieve results that used to be generally attainable in the past. Having an abstraction layer with a visualization layer will enable businesses to quickly and effectively respond to the rapidly changing global marketplace of the 21st century.
John Connor is chief technology officer at ASG. He has 30 years of experience in automation, help desk implementation, operations productivity, and enterprise-systems management in large information center environments. He is a frequent speaker at national and international industry events, such as Gartner, Forrester, IDC, SHARE, CMG, Guide, and AFCOM. You can contact the author at [email protected].