We present a novel approach to comprehending the behavior of distributed systems through the lens of cognitive linguistics. We introduce sixteen signs that encapsulate the fundamental operations and outcomes in service interactions. These signs, along with source information and orientation (release/receipt), form signals—the complete utterances of service communication. Unlike conventional metrics that provide isolated measurements, this approach treats service behavior as a language where meaning emerges from the sequential arrangement of signs. We illustrate how services exhibit varying levels of cognitive capability through their behavioral repertoire—the subset of signs they can express. This spectrum ranges from basic services with simple lifecycle signs to sophisticated services capable of complex adaptation and resource management. By adopting this linguistic framework, we enhance system observability and gain a deeper understanding of how distributed systems reason and adapt to their environment. This cognitive model of service behavior offers novel insights into system design, monitoring, and the evolution of distributed system intelligence.
- Service Cognition – Quantitative to Qualitative
- AIOps – SignOps + TaskOps
- A General Theory of System Observability
- Semiosphere: A Foundation for Perception and Control
The future of observability is all about creating new tools and taxonomies that let us understand how systems work at various scales and across state spaces. Companies that invest in extensible observability platforms get a big edge by learning more about their systems, which leads to making them faster, more reliable, and exceedingly effective. We must radically rethink observability, moving beyond the yesteryear three-pillar approach.
- Observability X – Location Agnostic
- Observability X – eXtensibility
- Observability X – Staging State
- Observability X – Naming Percepts
- Observability X – Contexts
- Observability X – Sources
- Observability X – Pipes & Pathways
- Observability X – Channels
- Observability X – Circuits
- Observability X – Subjects
- Observability X – States and Slots
- Observability X – Resources, Scopes, and Closures
- Observability X – Queues, Scripts, and Currents
- Observability X – Composers
- Observability X – Substrates 101
- Observability X – Containers
In this age of AI, how can we ensure that humans and AI work together as best as they can? The answer is to change how we think about working with AI. Instead of thinking of AI as a whole, let’s focus on its specific tasks. By doing this, we can bridge the gap between humans and machines and ensure they work together seamlessly.
In computing, a perpetual struggle ensues between the allure of simplicity and the inescapable progression towards complexity. Engineers and designers envision crafting elegant, uncomplicated solutions with intuitive interfaces, seamless workflows, and user-friendly interactions. However, beneath these aesthetically pleasing facades, a vast network of interconnected systems, microservices, and dependencies expands exponentially. This phenomenon mirrors the concept of entropy, as complexity appears to permeate our technological landscape regardless of our efforts to mitigate it. This raises questions regarding the nature of progress. Are we destined to construct increasingly complex systems to address more intricate challenges, or can we discover a path that harnesses complexity while maintaining simplicity?
- Climbing the Conceptual System
- From Abstraction to Simplicity
- Systems, Silos, and Simplicity
- The Complexity of Simplification
- Simplicity is Intelligence
The push for faster “real-time” feedback loops between software machines and man will result in the projection of both their behaviors into the same simulated universe. Within this simulation, a typical behavioral model consisting of actors, activities, and resources will unify both worlds sufficiently, allowing a business to monitor and manage operations oblivious to the actual nature of an actor.
- The Data Fog of Observability
- Beyond Big Data – Mirrored Algorithmic Simulation
- Transcending Code, Containers, and Cloud
- The Past, Present, and Future will be Simulated
- Papers We Love: The Past, Present, and Future Will Be Simulated (2014)
We firmly believe that Signals holds the potential to revolutionize the design and development of software, the performance engineering of systems, and the management of distributed interconnected applications and services. We initially designed Signals to analyze the sub-microsecond execution performance variations of software in extremely low-latency, high-frequency transaction and messaging environments. Later our design evolved to support the development of software components and libraries with self-adaptive capabilities.
- Simplicity and Significance in Observability
- The Evolution of Substrates
- A Story of Observability
- Observability in Perspective
- The Data Fog of Observability
- Observability – The Significant Parts
- Introducing Signals – The Next Big Thing
It’s hard to say exactly what AIOps means right now since it’s still a new and evolving concept in the IT world. However, most people agree that the goal of AIOps is to help humans manage complex software systems and microservices more efficiently. A key goal of operational intelligence, whether artificial or not, is to accurately assess whether the system has and is currently operating reliably and to determine whether an intervention in the form of change (actuation, configuration) is required when the assessment isn’t favorable, or there’s a prediction that things (situation) are about to change adversely. HAL 9000!
- The Intelligence in AIOps
- AIOps – The Observer
- AIOps – Why Service Cognition?
- AIOps – Visibility and Cognition
- AIOps – The Double Cone Model
- AIOps – A Postmodern Observability Model
- Streamlining Observability Pipelines
- Climbing the Conceptual System
We’ve got a fresh idea for improving site engineering reliability and service operations. We’re proposing a model that sets the stage for developing situational awareness and system resilience. This model focuses on adaptability and experimentation, which are key to dealing with dynamic configurations and chaos engineering. Our framework is clear and concise, and it’s designed to create effective and efficient layered cognitive structures and processes that allow machines and humans to work together seamlessly.
- Measurement and Control 2022
- AIOps – The Observer
- AIOps – A Postmodern Observability Model
- Observability – A Multitude of Memories
- Humanizing Observability and Controllability
- Scaling Observability for IT Ops
- Observability – The Two Hemispheres
- Situational Awareness in Systems of Services
- Observability: The OODA Loop
- Observability: Projecting Ahead
- Observability: Disruptions
- A Situational Control Tower
- Change in Observability
- The Data Fog of Observability
- Bounded Observability
- Service Cognition – Quantitative to Qualitative
- AIOps – SignOps + TaskOps
- Observability is about (to) Change
- The Situation Room of the Known Unknown