Micro Focus is now part of OpenText. Learn more >

You are here

You are here

Optimize Multicloud Ops Through Pragmatic Observability

David Linthicum Chief Cloud Strategy Officer, Deloitte Consulting
Photo by Michael Dziedzic on Unsplash

As growing numbers of organizations have embraced the deployment of hybrid and multicloud architectures, the issue of complexity is presenting a dilemma. On the one hand the functions and technologies those deployments bring can be vital to support productivity and innovation. On the other hand, many enterprises quickly reach a "complexity wall," where the complication added with each new technology threatens to diminish the returns on those strategic investments.

For all of their business value, best-of-breed technologies can have a significant downside. To maintain their competitive advantage, business leaders need to formulate approaches and innovations to manage complexity for both current and future systems—and they need to do so in a scalable way.

The key to achieving this is observability. Operations teams that embrace observability can break through the complexity wall.

Help Is on the Way

To this end, I've written a guide to help businesses create a plan to manage the growth of operational complexity associated with hybrid and multicloud architectures. This guide, Pragmatic Observability Trends and Techniques to Optimize Multicloud Operations, is designed to provide a better understanding of the problem and to offer pragmatic guidance to win the complexity battles before they negatively impact the business.

The conclusions reached in this guide include the following:

  • Complexity is a reality that can't be avoided.
  • Complexity results from leveraging best-of-breed technologies that use core cloud services optimized for a specific solution.
  • The primary weapons to deal with complexity are automation and abstraction tools that leverage the objectives of observability.
  • Organizations need to move from abstract insights to practical and immediately actionable insights.
  • Observability needs to be a pragmatic concept that creates objectives for both the cloud and legacy systems that comprise today's IT operations.
  • The success of these solutions draws from the capabilities and creativity they enable to deliver insights from the gigabytes of data that spin off traditional monitoring systems and emerging AIOps platforms.
  • Transparency of trust is a core issue.
  • Success requires identification of the most likely changes required in operational processes and tooling.

The observability maturity models presented in this guide will help determine your organization's level of observability maturityand what can be done to progress to higher maturity levels that provide more business value.

Cloud Complexity and Costs Are the Problem

Innovation can be beneficial, but it frequently leads to added complexity. In their quest to maximize productivity and facilitate innovation using best-of-breed services, developers will often deploy more and more different platforms across the environment. This naturally leads to a greater number of endpoints to operate—and, with them, added levels of complexity to manage. Thus, the likelihood of outages and data breaches increases.

Recent research indicates that public cloud outages are indeed getting worse. According to the Uptime Institute's 2022 outage analysis report, while outage rates in general have remained about as high as they ever were, the prevalence of major outages has shown a slight upward trend over the past three years. Moreover, one in five organizations reported experiencing a "serious" or "severe" outage—in other words, an outage that resulted in significant financial losses, reputational damage, compliance breaches, or even loss of life.

Data breaches, too, have consistently been on the rise. According to the Identity Theft Resource Center, data compromises reached an all-time high last year. And in its 2022 Cost of a Data Breach report, IBM reported that 83% of organizations it surveyed had experienced more than one data breach in their lifetime. IBM also found that nearly half of all data breaches involved the cloud.

Complexity is a natural byproduct of multicloud and other complex distributed architecture deployments. Given the innovative advantages that they offer, these types of deployments cannot be avoided. And while a complexity-induced rise in data-breach or cloud-outage risk could conceivably be offset by doubling or tripling operations budgets and hiring skilled staff, this is not a viable option for most corporate leaders.

Observability to the Rescue?

So what can be done to deal with complexity surges? Some industry pundits suggest that the answer lies in providing yet another layer of technologies and approaches. They promote buzzwordy ideas such as observability and AIOps as the solution to rising complexity insofar as these things can improve application monitoring and provide better insights into (and automation of) traditional IT systems.

The reality, however, is very different from the premise. Operations teams that adopt observability tools often discover a gap between these tools' value propositions and team members' ability to use them effectively.

This is largely about misunderstandings of the nature of observability from a pragmatic, practical sense—and how it differs from monitoring. Whereas monitoring refers to something you do, observability is a system attribute. It is the measure of how well internal system states can be inferred from knowledge of external data and external system states—in other words, how observable it is. As such, observability requires a holistic understanding of the deeper meaning of operational data.

To wit, if you can't make accurate assessments of your internal system states with all the external data in the world, then it doesn't matter how much monitoring you do. For instance, we may be able to see that our database system is running and showing a green status. But we may not see that our database is about to saturate allocated storage and corrupt the database. While monitoring is important, it is tactical. Observability, on the other hand, is strategic.

The new push by the cloud-operations industry is toward the more pragmatic application of observability as it relates to operations. In this context, the push for observability includes normalizing approaches and concepts in order to make them more relatable and actionable by rank-and-file operations teams. This means finding more pragmatic uses for observability concepts and devising better ways to present the insights that advanced operations tools can provide.

    To maximize the ROI of the entirety of their technology investments, IT leaders need to reframe their observability goals and approaches for cloud and IT operations. Given the rapid increase in operational complexity, organizations need to create change plans that consider the impact on culture, processes, approaches, and the technologies that make up their environment.

    Call to Action

    Today's businesses require actionable insights. When it comes to the optimization of IT operations, organizations can leave money on the table when they use solutions that do no more than provide tactical metrics. The biggest opportunities to recoup that money lie in the optimization of existing legacy systems and emerging multicloud deployments. IT organizations that do not track the ROI of their operations can easily overlook these opportunities.

    The operational concept of observability, combined with the use of parallel emerging technologies (such as AIOps), will help take operations to the highest maturity levels—optimizing operations' ability to return more value to the business. As such, this approach to observability acts as a true force multiplier.

    Business leaders therefore need to understand the core tenets of observability. They also need to learn how to weaponize observability using AIOps and other tools that can maximize business agility and innovation. Businesses that ignore observability run the risk of operations that fail to provide business value—and, possibly, drive business failure.

    Download the guide Pragmatic Observability Trends and Techniques to Optimize Multicloud Operations now.

    Keep learning

    Read more articles about: Enterprise ITIT Ops