The Financial and Business Impacts of an Ineffective IT Infrastructure
Discover the direct and indirect costs of IT infrastructure that doesn't meet your organization's needs.
By John F. Ryan04/09/2020
While most businesses are attempting to leverage the newest developing trends and innovations in IT computing (e.g., hybrid cloud, blockchain, containers, pervasive encryption, etc.), many are still struggling to meet their day-to-day SLAs. The recipe for successfully leveraging these newest innovations, starts with an IT infrastructure foundation that’s effective and operationally sound.
In this article, we'll examine the major differences between an effective and ineffective IT infrastructure and how these factors can have important cost and business ramifications.
Looking for Indicators
When we can't get a car to start, the first question someone might ask is “What are the indicators of trouble?” The same is true when examining the stability of any IT infrastructure. Figure 1 (below), outlines some of the key indicators of ineffective IT environments.
An effective IT infrastructure has cost structures that are easy to identify, manage and control. This encompasses all elements of cost, including hardware, software, maintenance, subscription support, labor, disaster recovery and facilities (power/cooling/space). This may seem straightforward, but in environments where there are many vendor server technologies1 in use, running applications from dozens to hundreds of different software vendors, all being managed by a variety of IT departments requiring different skills, the complexity in managing and controlling any related costs is greatly amplified.
The Impact of Outages
No one can predict unplanned outages. The adage “Plan for the worst, hope for the best” is certainly sound advice. IT shops with effective infrastructures tend to have this tenet down pat. They traditionally exceed most, if not all, aspects of five nines (99.999%) of availability and recoverability in order to meet the business’s required SLAs. This typically includes a robust, well planned, localized high availability as well as remote disaster recovery set of systems resources, processes and personnel appropriate for the task. Nevertheless, it’s not surprising to find that third-party organizations like Information Technology Intelligence Consulting (ITIC) and Ponemon Institute continue to find many ineffective infrastructures experiencing very debilitating outages costs.
Outages from security exposures or events are a specialized variation on the previous theme of maintaining superior levels of uninterrupted service. Businesses with effective IT infrastructures have integrated security (and all that implies) into their everyday practices and processes. Businesses with ineffective infrastructures tend to take an “additive approach” by slapping on a security package or widget as an after-thought. The result tends to be a more complex set of tools and processes, making interoperability and compatibility more difficult to manage.
Flexibility and agility are often underrated when discussing IT infrastructure effectiveness. Yet of all of the elements outlined here, they’re among the most important. An inflexible IT infrastructure doesn’t support a company’s need to respond to rapid course changes in its business goals or computing needs. This could be as simple as the ability to change compute capacity based on seasonal or intermittent business demands. Or it could be as complex as the need to define and support a corporate-wide standard for open container development. Whatever the need, an effective infrastructure is better suited to adjusting to these changes as they occur.
Know Your Costs: Direct Versus Indirect
Although many businesses focus mightily on negotiating the best discount they can when making a new IT infrastructure acquisition, organizations rarely know their total cost of operation. Reasons for this dearth of data include the range of suppliers and contracts used, irregular software billing methods, the spread of costs across many different lines of business, vagaries of asset ownership, the population of poorly documented systems, inconsistent reporting and auditing practices, and so on.
As was mentioned earlier, companies with effective infrastructures typically have a better understanding of the ongoing cost components of their IT operational needs. They also understand the distinction between direct and indirect costs.
Simply put, direct costs are those costs that can be directly tied to a specific area of use and support. Examples include enterprise mainframe, distributed UNIX*, distributed x86, desktop computing and storage. Indirect costs are those costs that cannot be tied to a specific area of use or support, but rather are tied to a set of resources or services used across the business. These indirect costs are usually split between allocated (shared services) and unallocated (overhead).
Without the ability to properly document and examine the direct and indirect cost equation, businesses lack proper respect for the various assets at their disposal. Ultimately, they run the risk of making poor decisions regarding any future investments in the effectiveness and welfare of their IT infrastructures.
Increasing IT Infrastructure Effectiveness
Figure 2 (above) comes from an actual study recently completed for a large telecommunications company. This business was experiencing major issues in meeting their performance and batch job SLAs. In addition, they had recently encountered a data security breach, which led to a two-hour disruption in service to their customers. Given the ineffective nature of the existing infrastructure, collecting pertinent inventory and cost data proved to be a significant challenge.
In examining their current environment, the IBM IT Economics team found significant opportunities to reduce infrastructure complexity and related costs by moving their key workloads from over 100 disparately managed systems to a single pair of IBM LinuxONE* servers. In particular, we found major savings potential in the areas of software subscription and support, systems administration, and disaster recovery.
The new IBM LinuxONE systems addressed any performance SLA issues by enabling the flexible use of Capacity-on-Demand. In addition, the encryption everywhere capability of LinuxONE offered the client a measure of certainty that any future risk of a data breach would be mitigated. Lastly, the team was able to prove that by rationalizing their existing infrastructure hodge-podge to a set of centrally managed, vertically scaled systems, the client would significantly improve their disaster recovery capabilities while reducing any associated costs.
The term “multiple” used here applies to multiple vendors, multiple types or architectures (x86, UNIX, IBM Z, appliance, etc.), multiple model configurations within a type and even multiple generations within a specific vendor model.
John F. Ryan is a senior certified IT technical specialist and consultant with IBM’s IT Economics Practice.