What is unified observability and why is it important? [Q&A]
There has been much discussion around observability in the past few months. With the adoption of hybrid work models and cloud systems, IT leaders have quickly realized the business and security value of creating transparency within their existing tech infrastructure.
Digital experience company Riverbed has announced a business move towards a market it's calling 'unified observability'. The company's vice-president Mike Marks spoke to us about the unified observability concept, why it's integral to IT decision-makers' strategies and offers some thoughts on how enterprises can begin integrating it within their current operations.
BN: There's been a lot of buzz around the monitoring, visibility and observability spaces. Could you outline what the differences and similarities are?
MM: As with any technology in the throes of the hype cycle, there's a lot of obfuscation and confusion as vendors try to fit themselves into the larger discussion.
Monitoring, visibility and observability are separate but complementary. Monitoring happens at the domain level and is symptom-oriented (e.g., slow application loading times), so it tells IT staff when something is wrong. Problems occur with monitoring programs because IT teams can quickly become overwhelmed with massive amounts of data or alerts if there isn’t any context.
Visibility is the next step up. Organizations have comprehensive monitoring in place and can aggregate and analyze data across domains. Visibility helps provide limited context to telemetry that monitoring often doesn’t point out.
Finally, observability is when visibility is paired with AI, ML and automation to provide actionable insights that help businesses understand unknowns, make better decisions, prioritize actions and solve problems faster.
BN: What's happening today to drive the need for unified observability?
MM: There are two key trends: relentless expectations around the user and customer digital experiences and excellent digital experiences are tied directly to organizational performance. Let’s break those down.
User expectations for an excellent digital experience are unrelenting, with customers and employees having a strong appetite for digital services. A recent Salesforce report found that 84 percent of people say a company's digital experience is as important as its products or services. This is especially challenging given that highly distributed, dynamic and hybrid environments, users, applications and data are everywhere. IT leaders have almost zero flexibility as, according to the Google Web team's RAIL performance model, users begin to lose focus on the task at hand after just a one-second delay. Perfection is now the norm.
When digital services fail to deliver a flawless experience, it undermines the businesses' ability to achieve essential objectives that drive revenue, reduce costs and remove risk. Digital performance is business performance. Top-notch application performance directly impacts productivity, leading to greater profitability and job satisfaction. No matter how invested businesses are in cloud and SaaS, sub-par app performance impacts productivity. Enterprises that can make good on that promise pave the way toward business benefits such as slashing their product's time to market, boosting supply chain agility and enhancing customer loyalty.
BN: What strategy is needed to help enterprises achieve unified observability?
MM: At Riverbed we've invested and focused the company's R&D efforts in the last year to develop technology and a unique approach to unified observability that is comprehensive, unified and easy to operate. CIOs will soon need to capture every packet, flow and end-user transaction across multi-cloud and on-premises networks, applications, and user systems at full fidelity. But they can't stop there. Organizations must also apply machine learning and artificial intelligence to help IT quickly identify and resolve digital service quality issues. Ultimately, observability without actionability doesn’t have much use in our fast-paced world.
Today's IT environments are interconnected, yet tech leaders often use siloed tools to manage performance and user experience. We see a world in which businesses leverage network performance management (NPM), IT Infrastructure Monitoring (ITIM) Application Performance Monitoring (APM) and Digital Experience Management (DEM) capabilities to meet the challenging demands of the modern business.
BN: If a CIO reading this wanted to begin working towards unified observability, where would they start?
MM: To deliver on employees' and customers' high expectations, organizations need a unified view, with intelligence that cuts through massive complexity to transform data into actional insights. The path towards unified observability isn't impossible or exclusive to well-resourced businesses. Many organizations may have some of the critical capabilities already in place at one stage or another.
Full-fidelity telemetry is the cornerstone of unified observability. Organizations must capture full-fidelity data from monitoring and visibility tools across the entire IT ecosystem, including client devices, networks, servers, applications, cloud-native environments, and users. This complete picture enables IT to understand what is happening and what has happened while not missing key events or context due to sampling.
AI and ML are also vital. Applying AI, ML, and data science techniques across multiple data streams, including third-party data, can better help IT teams detect anomalies and changes. The organization can surface the most critical issues faster and enables better prioritization to focus IT teams' time and effort on the most impactful projects and issues.
Image credit: galgogczygabriel/depositphotos.com