This content is part of the Essential Guide: An initiation into infrastructure automation tools and methods

Create a change management system to tame data center chaos

Data center change is constant, thanks to new systems, upgrades, user changes and macro trends like the introduction of mobile devices for business purposes.

IT organizations are in a constant struggle to manage changing data center configurations, often with an outdated change management system.

Holes in business processes and limitations with infrastructure management tools have put many businesses in a precarious position, which is only compounded as the pace of change in enterprise use of technology ramps up. Yet, flexibility and agility are not antithetical to system manageability.

System configuration management has grown from inefficient spreadsheets to automated programs. Traditionally, change management systems focused on alterations made to physical devices on the enterprise network, such as a server memory upgrade or new branch router. Modern change management works at the software level, monitoring such things as router OS patches or an upgrade to the finance department's accounts payable application.

In terms of IT systems, "flexible" and "agile" are two adjectives that resonate with upper management. Data center staff try -- sometimes unsuccessfully -- to enhance these capabilities without creating configuration confusion, introducing new system inefficiencies or compromising security. A data center that is in a constant state of flux is difficult to manage.

"IT is struggling to keep pace with the speed seen with mobile, social and cloud systems," stated Mary Johnston Turner, research vice president at International Data Corporation. Data center managers risk creating unforeseen problems with every upgrade to interconnected systems.

Configuration and change management (C&CM) systems are available either integrated into enterprise system management suites or as standalone products. Leading management suppliers, like BMC Software Inc., CA Technologies, Hewlett-Packard, IBM and Microsoft, offer the former; suppliers such as Axios Products Inc. and LANDESK Software specialize in the latter.

Tools can't keep pace with technology

The types of applications for workers are also growing. IT easily kept tight control over enterprise resource planning and other traditional systems. However, employees now are likely to buy new software, like mobile or social products, on their own.

A data center in a constant state of flux is difficult to manage.

In fact, market research company CCS Insight found that 80% of employees do not bother to ask IT for help with mobile apps, preferring to download them on their own. Four out of five employees in the survey told CCS Insight that mobile is important for doing their jobs. This means that a lot of software is used within the corporate walls that IT did not authorize and may not even know exists. The group has limited recourse -- in some cases none -- for reining in the user-deployed solutions.

As the pace of change has increased, the visibility tools -- and IT -- have fallen behind for various reasons. The connections among different systems are more complex and difficult to track. End users work on a wide array of devices, with mobile systems becoming as common as PCs.

Virtualization's change management tradeoffs

Virtualization has enabled companies to move information among a variety of VMs, which increases system efficiency but makes tracking what happens on those systems difficult. VMs do not operate in vacuums; their actions must be tied to system infrastructure solutions, such as authorization systems and licensing programs. Data center personnel put these connections in place and then maintain them as the system evolves.

With a poor change management system in place, connections are not made or fall out of date, so IT cannot track system usage. VMs are allocated dynamically. If a business is licensed for 100 users and fails to structure the licensing software to interact with the VM software, it could be noncompliant with access for 105 users.

In another case, a given VM may process the user interface for an account in one moment. Five minutes later, that same VM may generate the firm's biweekly paychecks. Payroll processing requires 2 Gb more random access memory than the user interface task. The virtualization system provisions the resource, but the alteration is not relayed to the change management system.

The personnel problem

Businesses need to put processes in place to capture rapidly changing system configurations. Each alteration reverberates throughout the organization, but employees focus on their own needs and do not recognize the ripple effects created by their actions.

One area where better communication is needed is between IT and HR. An Intermedia study found that 89% of ex-employees retain access to company applications, 49% log into closed accounts, and 45% access confidential data. IT organizations must feed relevant HR data into their change management systems, so access to company systems ends when employees leave the firm.

Big data, SDDC push management envelope

Similar challenges arise with data storage and networking as software-defined data centers take shape. SDDCs bring virtualization features to storage and enterprise networks. Analyst firm Research and Markets expects revenue for the SDDC market to increase from $21.78 billion in 2015 to $77.18 billion by 2020, a 28.8% compound annual growth rate. While businesses gain agility from SDDC, more dynamic resource allocation comes at the cost of system visibility.

The need for real-time data analytics is increasing, so big data functionality is being integrated into C&CM systems. In addition, Puppet, Chef and other open source orchestration tools are emerging. These tools automate and coordinate change processing across various infrastructure management systems.

Data center organizations are pushing vendors to deliver stronger change management solutions, and the established IT vendors as well as the specialized C&CM vendors struggle with these complex issues.

About the author:
Paul Korzeniowski is a freelance writer who specializes in data center issues. He has been writing about technology for two decades, is based in Sudbury, MA and can be reached at
[email protected].

Next Steps

Manage IT assets, reduce chaos and save money

Take the pain out of change management

Dig Deeper on Configuration Management and DevOps