Glossary

Agentic AI: Autonomous systems powered by Large Language Models (LLMs) that go beyond simple task execution to plan, reason, and execute complex goals using external tools, memory, and self-correction to complete multi-step workflows with minimal human intervention. See course-notes/google-cloud-digital-leader/digital-transformation.
Agile: An umbrella term for a collection of frameworks and techniques that together enable teams and individuals to work in a way that is typified by collaboration, prioritization, iterative and incremental delivery, and time-boxing. These are several specific methods (or frameworks) that are classed as Agile, such as Scrum, Lean, and Kanban.
AI Hypercomputer: Google Cloud's advanced, specialized infrastructure designed to train and deploy complex machine learning and AI applications with maximum performance, efficiency, and cost savings, featuring custom-designed TPUs and planet-scale networking. See course-notes/google-cloud-digital-leader/digital-transformation.
APIRouter: A FastAPI class used to organize and decouple endpoint routing. It allows developers to split large applications into multiple modular files, preventing monolithic architecture. See textbooks/fastapi/index for usage.
AutoML: A Google Cloud service that automates the machine learning model training and tuning process visually through a dashboard without writing code, comparing algorithms, and tuning parameters automatically. See course-notes/cloud/google-cloud-digital-leader/innovating-with-google-cloud-artificial-intelligence.
ARP Cache: The local table used by an operating system to map IP addresses to physical MAC addresses. Following Keepalived or Kube-VIP failovers, stale entries in this cache on client machines can lead to connection timeouts (e.g. connect: no route to host). See how-to/post-power-outage-recovery.
acme-companion: A companion Docker container for nginx-proxy that automates the creation, renewal, and installation of Let's Encrypt SSL certificates by monitoring Docker events and verifying domains via HTTP-01 challenges. See how-to/host-multiple-docker-projects-with-shared-nginx-proxy.
Bandwidth: The measure of how much data a network can transfer in a given amount of time, typically measured in Megabits (Mbps) or Gigabits (Gbps) per second. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.
Build-time Serialization: The build-time process of scanning, parsing, and converting dynamic file-based resources (such as Markdown files) into a static, optimized payload (like JSON). This technique decouples the physical content repository from the production backend environment, maximizing security and eliminating runtime parsing overhead. See how-to/build-zettelkasten-graph-fastapi for practical details.
Breakpoint (Responsive): A specific screen width threshold defined in CSS media queries (e.g., @media (max-width: 1023.98px)) that triggers layout and styling changes to adapt the user interface to different devices. See textbooks/fastapi/index.
change: The addition, modification, or removal of anything that could have a direct or indirect effect on services.
change authority: The person or group who authorizes a change.
change schedule: A calendar or database used to plan changes, facilitate communication, avoid conflicts, and allocate resources.
CI: (Configuration Item) Any component that needs to be managed in order to deliver an IT service.
cordon: (Kubernetes) Marking a node as unschedulable so that the Kubernetes scheduler will not place any new pods on it. Existing pods continue running. Used before maintenance operations. See also: drain, uncordon.
CIR: (Continual Improvement Register) A structured database or document that is used to track and manage improvement opportunities.
Circular Import Mitigation: The software engineering practice of resolving cyclic dependencies between modules (e.g., Module A importing Module B while Module B imports Module A). Common techniques include moving shared instantiation logic to a third, independent module or utilizing local imports inside functions. See textbooks/fastapi/index for a theoretical breakdown.
Cache-Buster: A mechanism used to force browser clients to bypass their local cache and fetch the latest version of a static asset (such as CSS or JS). This is achieved by appending a unique query parameter signature (e.g., ?v=1.0.1) to the asset URL, altering its cache key. See learning/fastapi/static-files.
Cache Layer (Docker): An intermediate read-only image layer generated during the execution of an instruction in a Dockerfile. Subsequent builds reuse these cached layers to accelerate build times, unless forced to compile cleanly using the --no-cache flag. See how-to/deploy-docker-compose-updates.
Container Recreation: The process of destroying a running Docker container and launching a fresh instance built from an updated image. Simply rebuilding an image does not automatically apply modifications to a running container; recreation is required (e.g. via docker compose up -d). See how-to/deploy-docker-compose-updates.
Colima: A lightweight, command-line container runtime for macOS (and Linux) that serves as an open-source alternative to Docker Desktop. It provisions a minimal Linux VM to run the Docker daemon. See how-to/setup-local-dev-docker-colima.
conntrack: A core Linux utility used by iptables to track network connections. It is a strict prerequisite for Kubernetes (kube-proxy) to function, and must be installed manually on minimal Debian netinst configurations.
continual improvement practice: The practice of aligning an organization's practices and services with changing business needs through the ongoing identification and improvement of all elements involved in the effective management of products and services.
Compute: In the cloud, compute describes the ability of a machine to process information, handling tasks like storage, retrieval, and analysis. Unlike traditional local servers, cloud computing uses a network of remote servers to offer on-demand access to resources. See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
cost: The amount of money spent on a specific activity or resource. Cost can also be expressed in non-monetary terms.
CRI: (Container Runtime Interface) The plugin interface which enables kubelet to use a wide variety of container runtimes, without the need to recompile Kubernetes.
CSF: (critical success factor) A necessary precondition for the achievement of intended results.
customer: The role that defines the requirements for a service and takes responsibility for the outcomes of service consumption.
CX: (customer experience) The sum of functional and emotional interactions with a service and service provider as perceived by a customer.
D3.js Force Simulation: A physics engine provided by the D3.js library that models physical forces (like charge, gravity, and link distance) between nodes and connections. It dynamically calculates coordinate positions in the client's browser to render interactive force-directed graph visualizations. See how-to/build-zettelkasten-graph-fastapi for practical details.
Database Portability: The architectural practice of designing database connections to adapt dynamically to the environment (e.g. falling back to a local zero-dependency SQLite file for local testing and connecting to a PostgreSQL container when running under Docker). See textbooks/fastapi/index.
Data Drift: The phenomenon where the statistical properties of real-world production data change over time compared to the dataset the machine learning model was trained on, causing performance degradation. See course-notes/cloud/google-cloud-digital-leader/innovating-with-google-cloud-artificial-intelligence.
Data Transformation: The process of capturing, managing, and using data to redefine customer experiences, create new value, and make smarter decisions. See course-notes/cloud/google-cloud-digital-leader/exploring-data-transformation-with-google-cloud.
Data-Value Gap: The challenge of bridging the divide between simply storing data and generating business value or predictions from it. See course-notes/cloud/google-cloud-digital-leader/exploring-data-transformation-with-google-cloud.
Decorator: A Python syntax construct that modifies the behavior of a function, method, or class. In web frameworks like FastAPI, decorators are used to bind URL routes to handler functions. See textbooks/fastapi/index.
demand: Input to the Service Value System based on opportunities and needs from internal and external stakeholders.
deployment management: The practice of moving new or changed hardware, software, documentation, processes, or any other service component to live environments. See course-notes/itil/index.
Dependency Deadlock: A scenario where service components fail to boot successfully because they depend on other services that are either offline or scheduled to start after them (such as Kubernetes pods crashlooping because the dynamic NFS storage provider cannot reach the offline NAS array). See how-to/post-power-outage-recovery.
DevOps: An organizational culture that aims to improve the flow of value to customers. DevOps focuses on culture, automation, Lean, measurement, and sharing (CALMS).
Digital Transformation: The strategic process whereby an organization uses modern digital technologies—including public, private, and hybrid cloud platforms—to create or modify business processes, culture, and customer experiences to meet changing business and market dynamics. See course-notes/google-cloud-digital-leader/digital-transformation.
Discovery and Assessment: The initial phase of a migration project where an organization inventories its existing on-premises environment (looking at applications, dependencies, and costs) to determine which workloads should move to the cloud and which migration strategy is appropriate. See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
.dockerignore: A configuration file used in Docker builds to specify files and folders (such as local .venv or temporary database directories) that should be excluded from the build context, preventing host environment pollution from corrupting the container filesystem. See how-to/host-multiple-docker-projects-with-shared-nginx-proxy.
DKMS: (Dynamic Kernel Module Support) A framework that enables generating Linux kernel modules whose sources reside outside the kernel source tree. Commonly used for proprietary drivers like NVIDIA.
drain: (Kubernetes) The process of safely evicting all pods from a node and marking it as unschedulable (cordoned). Used to prepare a node for maintenance, upgrades, or shutdown. See also: cordon, uncordon.
Dynamic Workload Scheduler: A Google Cloud resource orchestration scheduler that minimizes costs on the AI Hypercomputer by managing capacities and provisioning resources exactly when needed for critical runs. See course-notes/cloud/google-cloud-digital-leader/innovating-with-google-cloud-artificial-intelligence.
event: Any change of state that has significance for the management of a service or other configuration item (CI).
ethtool: A Linux utility for controlling network driver and hardware settings. It is required by kubeadm during the Kubernetes cluster bootstrap preflight checks.
Explainable AI: A suite of tools and frameworks designed to interpret and explain machine learning model predictions to ensure transparency and build user trust. See course-notes/cloud/google-cloud-digital-leader/innovating-with-google-cloud-artificial-intelligence.
FastAPI: A modern, high-performance web framework for building APIs with Python 3.8+ based on standard Python type hints. Features automatic OpenAPI documentation, automatic validation, and serialization. See textbooks/fastapi/index.
feedback loop: A technique whereby the outputs of one part of a system are used as inputs to the same part of the system.
File-Watching VM Barrier: The boundary in virtualized development environments (like Docker containers running inside a macOS hypervisor VM) where host filesystem modification events fail to propagate to the guest OS. This requires configuring watchers to actively poll directories. See textbooks/containerization/index.
four dimensions of service management: The four perspectives that are critical to the effective and efficient facilitation of value for customers and other stakeholders in the form of products and services.
Function: A reusable block of code designed to perform a specific task, taking input arguments and returning outputs.
Gemini Enterprise Agent Platform: Google Cloud's unified platform (including Agent Studio and notebooks) for building, deploying, tuning, and managing custom and foundational AI models. See course-notes/cloud/google-cloud-digital-leader/innovating-with-google-cloud-artificial-intelligence.
Generative AI: A type of artificial intelligence capable of producing new content—including text, images, audio, video, and code—by learning from existing training data. See course-notes/cloud/google-cloud-digital-leader/innovating-with-google-cloud-artificial-intelligence.
Google Edge Network: The entry point where a device or local ISP connects to Google's private network, utilizing cached content and optimized routing paths to deliver services with high reliability and low latency. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.
governance: The means by which an organization is directed and controlled.
holistic: The treatment of the system as a whole and not just the sum of its parts.
HBA: (Host Bus Adapter) A circuit board and/or integrated circuit adapter that provides I/O processing and physical connectivity between a host system and storage devices. Often flashed to "IT Mode" for software RAID (ZFS).
Hybrid Cloud: An IT deployment strategy that combines public and private cloud environments (such as an on-premises data center and Google Cloud) to run applications and share data across them. See course-notes/google-cloud-digital-leader/digital-transformation.
IaaS (Infrastructure as a Service): A cloud computing service model that offers virtualized computing resources (compute, storage, networking) over the internet on a pay-as-you-go basis, where the customer manages the operating system, middleware, and applications. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.
incident: An unplanned interruption to a service, or reduction in the quality of a service.
information and technology: One of the four dimensions of service management. It includes the information and knowledge used to deliver services, and the information and technologies used to manage all aspects of the Service Value System.
Integration Tax: The hidden, non-monetary costs and drawbacks (paid in complexity, high latency, new security risks, and runaway costs) incurred when attempting to run modern AI workloads on top of legacy, fragmented, or proprietary cloud architectures. See course-notes/google-cloud-digital-leader/digital-transformation.
Intelligent Data Cloud: A unified platform that integrates all data types, operates at a real-time scale, and embeds machine learning and AI directly into application workflows to close the data-value gap. See course-notes/cloud/google-cloud-digital-leader/exploring-data-transformation-with-google-cloud.
IT asset: Any financially valuable component that can contribute to delivery of an IT product or service.
ITIL: Best-practice guidance for IT service management.
ITIL guiding principles: Recommendations that can guide an organization in all circumstances, regardless of changes in its goals, strategies, types of work, or management structure.
ITIL service value chain: An operating model for service providers that covers all the key activities required to effectively manage products and services.
JBOD: (Just a Bunch Of Disks) A storage architecture using multiple hard drives exposed directly to the operating system as individual disks, allowing software like ZFS to manage them instead of a hardware RAID controller.
Kanban: A method for visualizing work, identifying potential blockages and resource conflicts, and managing work in progress.
known error: A problem that has been analyzed but has not been resolved.
lazydocker: A terminal-based user interface (TUI) dashboard for managing Docker containers, images, volumes, and networks. See how-to/setup-local-dev-docker-colima.
KPI: (key performance indicator) An important metric used to evaluate the success in meeting an objective.
Kube-VIP: A lightweight Kubernetes load-balancing tool that provides a highly available virtual IP (VIP) address for control plane nodes, managing VIP bindings dynamically to prevent API server access loss. See how-to/post-power-outage-recovery.
Latency: The time it takes for a single packet of data to travel from its source to its destination, typically measured in milliseconds (ms) and heavily influenced by physical distance. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.
Lean: An approach that focuses on improving workflows by maximizing value through the elimination of waste.
Machine Learning: A subset of artificial intelligence that allows computer systems to learn from data and improve performance over time without being explicitly programmed. See course-notes/cloud/google-cloud-digital-leader/innovating-with-google-cloud-artificial-intelligence.
metric: A measurement or calculation that is monitored or reported for management and improvement.
Model Garden: A unified repository within Agent Studio for discovering and selecting Google foundation models and popular open-source models. See course-notes/cloud/google-cloud-digital-leader/innovating-with-google-cloud-artificial-intelligence.
monitoring and event management: The practice of systematically observing services and service components, and recording and reporting selected changes of state identified as events. See course-notes/itil/index.
MIME Sniffing: A mechanism by which web browsers examine the content of a byte stream to deduce the file type, rather than relying strictly on the server-provided Content-Type header. To mitigate security vulnerabilities, this behavior can be blocked using the X-Content-Type-Options: nosniff header, which forces the browser to obey the server's declared MIME type.
Mixed Content: A security mechanism where web browsers loading a page over a secure connection (https://) block active resources (such as CSS stylesheets or script files) requested over an insecure connection (http://) to prevent interception and manipulation. See reference/web-development/fastapi-static-files.
mission: The overall purpose and intentions of an organization.
Multicloud: An IT infrastructure architecture that combines services and resources from at least two public cloud providers (such as Google Cloud, AWS, and Microsoft Azure) to utilize their respective strengths and avoid vendor lock-in. See course-notes/google-cloud-digital-leader/digital-transformation.
Multi-region (Cloud): A cloud deployment model where data and resources are stored redundantly in at least two geographic locations separated by a minimum distance (e.g., 160 kilometers) to ensure high durability and disaster recovery. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.
MVP: (Minimum Viable Product) A product with just enough features to satisfy early customers, and to provide feedback for future product development.
node: (Kubernetes) A machine (physical or virtual) that is part of a Kubernetes cluster. Control-plane nodes manage the cluster; worker nodes run application workloads.
NFS: (Network File System) A distributed file system protocol allowing a client computer to access files over a network much like local storage. Used in Kubernetes for ReadWriteMany persistent volumes.
Negative Margin (CSS): A styling technique where negative values are applied to an element's margin (e.g., margin-left: -2.5rem), pulling the element outward beyond its parent container's standard padding bounds. Commonly used to create full-bleed elements or expand nested structures on mobile screens. See textbooks/fastapi/index.
Nginx Reverse Proxy: A server that sits in front of backend applications and forwards client requests to those applications based on rules like domain names. Commonly used to host multiple Docker containers on the same server sharing ports 80 and 443. See how-to/host-multiple-docker-projects-with-shared-nginx-proxy.
only-system (Python Preference): A configuration setting in the uv package manager (UV_PYTHON_PREFERENCE=only-system) that forces it to resolve dependencies against the host or container's system-installed Python interpreter, preventing it from downloading standalone python runtimes that would create broken virtualenv symlinks in multi-stage builds. See how-to/host-multiple-docker-projects-with-shared-nginx-proxy.
Open Source: Software whose source code is publicly accessible and free for anyone to use, modify, and share, developed via public collaboration to provide portability and prevent vendor lock-in. See course-notes/google-cloud-digital-leader/digital-transformation.
Open Standards: Software specifications and guidelines that are openly accessible and usable by anyone (e.g., HTTP, XML), ensuring products perform in an interoperable way and preventing vendor lock-in. See course-notes/google-cloud-digital-leader/digital-transformation.
opportunity: Options or possibilities to add value for stakeholders or otherwise improve the organization.
organization: A person or a group of people that has its own functions with responsibilities, authorities, and relationships to achieve its objectives.
organizations and people: One of the four dimensions of service management. It ensures that the way an organization is structured and managed, as well as its roles, responsibilities, and systems of authority and communication, is well defined and supports its overall strategy and operating model.
outcome: A result for a stakeholder enabled by one or more outputs.
output: A tangible or intangible deliverable of an activity.
PaaS (Platform as a Service): A cloud computing service model that provides developers with a pre-configured platform (hardware, operating systems, runtime, and middleware) to build, run, and manage custom applications without the complexity of maintaining the underlying infrastructure. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.
partners and suppliers: One of the four dimensions of service management. It encompasses the relationships an organization has with other organizations that are involved in the design, development, deployment, delivery, support, and/or continual improvement of services.
partnership: A relationship between two organizations that involves working closely together to achieve common goals and objectives.
performance: A measure of what is achieved or delivered by a system, person, team, practice, or service.
pod: (Kubernetes) The smallest deployable unit in Kubernetes. A pod wraps one or more containers and provides them with a shared network namespace and storage volumes.
practice: Sets of organizational resources designed for performing work or accomplishing an objective.
Pre-trained APIs: Out-of-the-box machine learning models developed and trained by Google (e.g., Speech-to-Text, Cloud Vision) that developers can plug into their applications without ML training or expertise. See course-notes/cloud/google-cloud-digital-leader/innovating-with-google-cloud-artificial-intelligence.
Private Cloud: A single-tenant cloud computing model where virtualized server infrastructure is dedicated exclusively to a single organization (corporate cloud), accessed over the internet but hosted within an organization’s own data center, a third-party colocation facility, or by a private cloud provider. See course-notes/google-cloud-digital-leader/digital-transformation.
Privilege Escalation: The process of obtaining elevated permissions (usually root or Administrator) on a computer system. In automation like Ansible, this is often handled by the become system using methods like sudo or su. See reference/ansible-privilege-escalation.
problem: A cause, or potential cause, of one or more incidents.
process: A set of interrelated or interacting activities that transform inputs into outputs. Processes define the sequence of actions and their dependencies.
product: A configuration of an organization's resources designed to offer value for a consumer.
Proxy Headers: HTTP request headers (such as X-Forwarded-Proto, X-Forwarded-For, and X-Forwarded-Port) injected by reverse proxies to convey the original client connection properties to the backend application server. See reference/web-development/fastapi-static-files.
psycopg: A modern PostgreSQL database adapter library for Python, conforming to the PEP 249 DB-API standard. See textbooks/fastapi/index.
Public Cloud: A multi-tenant cloud computing model where on-demand computing services and infrastructure are managed by a third-party provider (such as Google Cloud) and shared among multiple organizations over the public internet, with each tenant's resources locked and private. See course-notes/google-cloud-digital-leader/digital-transformation.
Refactoring: The process of changing the code of a workload, such as transitioning it to a cloud-based microservices or serverless architecture, to make it more efficient, scalable, or secure. See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
Region (Cloud): An independent geographic area (e.g., London, europe-west2) composed of multiple zones, serving as a deployment location for cloud resources. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.
Rehosting: The migration of a workload to the cloud without changing anything in the workload's code or architecture (often referred to as "lift and shift"). See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
Reimagining: The process of rethinking how an organization uses technology (such as artificial intelligence and machine learning) to achieve its business goals. See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
release: A version of a service or other configuration item, or a collection of configuration items, that is made available for use.
Replatforming: The process of migrating a workload to the cloud while making minor changes to its code or architecture to leverage cloud-native scalability and reliability (often called "move and improve"). See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
Request Body: JSON (or other format) data transmitted within the body of an HTTP request (such as POST or PUT) to an API. In FastAPI, request bodies are typically defined and validated using Pydantic models. See textbooks/fastapi/index.
resource: Personnel, material, finance or other entity required for the execution of an activity or the achievement of an objective. Resources used by an organization may be owned by the organization or used according to an agreement with the resource owner.
Response: The payload or data returned by an API endpoint to the client after processing a request (e.g. JSON, HTML, plain text, or binary files). See textbooks/fastapi/index.
Restart Policy (Docker): A configuration directive in Docker Compose (restart: always or restart: unless-stopped) that instructs the engine to automatically restart containers when they exit, crash, or when the host daemon restarts. See how-to/host-multiple-docker-projects-with-shared-nginx-proxy.
Restore on AC Power Loss: A motherboard UEFI/BIOS power management setting that determines whether a physical server automatically boots back up when utility power is restored after a blackout. See how-to/post-power-outage-recovery.
Retaining: The intentional practice of keeping a workload on-premises or in a hybrid cloud environment, to be managed internally by the business rather than subject to cloud provider control. See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
Retiring: The removal of a workload from a platform because it is unnecessary, not cost-effective, not secure, or incompatible. See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
risk: A possible event that could cause harm or loss, or make it more difficult to achieve objectives. Risk can also be defined as uncertainty of outcome, and can be used in the context of measuring the probability of positive outcomes as well as negative outcomes.
Route: A URL path mapped to a specific handler function in a web application or API. In FastAPI, routes are declared using decorator syntax on the application or router instance. See textbooks/fastapi/index.
SaaS (Software as a Service): A cloud computing service model that delivers a complete, fully managed software application over the internet (typically via a web browser), requiring no installation or infrastructure management by the customer. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.
Secure Boot: A UEFI feature that prevents malicious software and unauthorized operating systems from loading during the boot process. It strictly blocks unsigned kernel modules (like DKMS proprietary drivers).
Semi-structured Data: Information that is not strictly tabular but contains metadata, tags, or markers (e.g. JSON, XML) to separate elements. See course-notes/cloud/google-cloud-digital-leader/exploring-data-transformation-with-google-cloud.
service: A means of enabling value co-creation by facilitating outcomes that customers want to achieve, without the customer having to manage specific costs and risks.
service catalog: Structured information about all the services and service offerings for a service provider, relevant for a specific target audience.
service consumption: Activities performed by an organization to consume services. It includes the management of the consumer's resources needed to use the service, service actions performed by users, and the receiving (acquiring) of goods (if required).
service desk: The point of communication between the service provider and all of its users.
service management: A set of specialized organizational capabilities for enabling value for customers in the form of services.
service offering: A formal description of one or more services, designed to address the needs of a target consumer group. A service offering may include goods, access to resources, and service actions.
service provider: A role performed by an organization in a service relationship to provide services to consumers.
service provision: Activities performed by an organization to provide services. It includes management of the provider's resources; configured to deliver the service, ensuring access to these resources for users; fulfillment of the agreed service actions; service level management; and continual improvement. It may also include the supply of goods.
service relationship: A cooperation between a service provider and service consumer. Service relationships include service provision, service consumption, and service relationship management.
service relationship management: Joint activities performed by a service provider and a service consumer to ensure continual value co-creation based on agreed and available service offerings.
service request: A request from a user or user's authorized representative that initiates a service action which has been agreed as a normal part of service delivery.
Shared Docker Network: An external Docker network that acts as a bridge, allowing completely independent Docker Compose projects to communicate securely with each other (such as a reverse proxy routing to separate backend stacks). See how-to/host-multiple-docker-projects-with-shared-nginx-proxy.
Shared Responsibility Model: A security framework that divides security obligations between the cloud provider ("security of the cloud") and the customer ("security in the cloud"), depending on the cloud service model used. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.
SLA: (Service Level Agreement) A documented agreement between a service provider and a customer that identifies both services required and the expected level of service.
sponsor: The role that authorizes budget for service consumption. Can also be used to describe an organization or individual that provides financial or other support for an initiative.
SSL Unrecognized Name Alert (SNI): A TLS warning alert (ERR_SSL_UNRECOGNIZED_NAME_ALERT) returned by a reverse proxy or web server during the handshake if a client requests a domain name (via SNI) that the server does not recognize or has no SSL certificate mapped for. See how-to/host-multiple-docker-projects-with-shared-nginx-proxy.
SSL Termination: The process of decrypting SSL/TLS encrypted traffic at a proxy, load balancer, or web server before passing the unencrypted HTTP traffic to the backend application server, relieving backend containers of cryptographic overhead. See how-to/host-multiple-docker-projects-with-shared-nginx-proxy.
Strict MIME-Type Checking: A security enforcement mechanism where web browsers refuse to load, execute, or apply external assets (such as CSS stylesheets or JavaScript files) unless the server transmits them with the precise, correct Content-Type header (e.g. text/css for style sheets). This security boundary prevents cross-site script execution and style injection. See learning/fastapi/static-files and how-to/configure-fastapi-static-mime for diagnostic steps.
Structured Data: Neatly organized, formatted information that easily fits into spreadsheets or relational tables (e.g. sales databases). See course-notes/cloud/google-cloud-digital-leader/exploring-data-transformation-with-google-cloud.
supplier: A stakeholder responsible for providing services that are used by an organization.
SVS: (Service Value System) A model representing how all the components and activities of an organization work together to facilitate value creation.
systems thinking: A holistic approach to analysis that focuses on the way that a system's constituent parts work, interrelate, and interact over time, and within the context of other systems.
TOC: (Theory of Constraints) A method of continual improvement that involves identifying the weakest link in the value chain (i.e. the constraint) and then making systematic improvements until it is no longer the weakest link.
Total Cost of Ownership (TCO): A financial metric that measures the total cost of a system over its entire lifetime, including purchase, operations, maintenance, and upgrade costs. Shifting from on-premises hardware to the cloud lowers TCO by replacing capital expenses with a pay-as-you-go operational model. See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
TPU: (Tensor Processing Unit) An application-specific integrated circuit (ASIC) custom-designed by Google from the ground up specifically to accelerate machine learning and AI workloads. See course-notes/google-cloud-digital-leader/digital-transformation.
uncordon: (Kubernetes) Marking a previously cordoned node as schedulable again, allowing the Kubernetes scheduler to place new pods on it. Typically performed after maintenance is complete and the node has rejoined the cluster. See also: cordon, drain.
Unstructured Data: Information without a pre-defined data model (e.g. social media posts, videos, audio, PDFs) that is highly valuable but harder to capture and analyze. See course-notes/cloud/google-cloud-digital-leader/exploring-data-transformation-with-google-cloud.
Uptime Monitor: An observability tool or service (such as Uptime Kuma or UptimeRobot) that periodically pings endpoints over HTTP/HTTPS/TCP to verify availability, latency, and SSL cert health, alerting operators of any failures. See how-to/post-power-outage-recovery.
user: The role that uses services.
utility: The functionality offered by a product or service to meet a particular need. Utility can be summarized as "what the service does" and can be used to determine whether a service is "fit for purpose." To have utility, a service must either support the performance of the consumer or remove constraints from the consumer. Many services do both.
vanilla CSS: Plain CSS code written directly in standard stylesheet files without using utility-first compilation frameworks (like Tailwind CSS) or processors (like Sass).
Variable Precedence (Ansible): The strict hierarchy rules Ansible uses to determine which variable wins when the same variable is defined in multiple places. For example, variables defined in the inventory always override those defined in the playbook. See reference/ansible-privilege-escalation.
UX: (user experience) The sum of functional and emotional interactions with a service and service provider as perceived by a user.
value: The perceived benefits, usefulness, and importance of something.
value stream: A step in the value chain that an organization takes in the creation of value.
value stream and processes: One of the four dimensions of service management. It defines the activities, workflows, controls, and procedures needed to achieve the agreed objectives.
vision: A defined aspiration of what an organization would like to become in the future.
visudo: A strict utility for editing the /etc/sudoers file. It locks the file against simultaneous edits and runs validation checks on save, preventing a malformed file from locking administrators out of root access.
warranty: Assurance that a product or service will meet agreed requirements. Warranty can be summarized as "how the service performs" and can be used to determine whether a service is "fit for use." Warranty often relates to service levels aligned with the needs of service consumers. This may be based on a formal agreement, or it may be a marketing message or brand image. Warranty typically addresses such areas as the availability of the service, its capacity, levels of security, and continuity. A service may be said to provide acceptable assurance, or "warranty," if all defined and agreed conditions are met.
Wikilinks: A shorthand hypertext syntax (typically formatted as [[Page Name]]) popularized by wikis and personal knowledge management tools (such as Obsidian). It allows direct cross-referencing between related articles or notes without specifying raw filesystem paths. See how-to/build-zettelkasten-graph-fastapi for practical details.
workaround: A solution that reduces or eliminates the impact of an incident or problem for which a full resolution is not yet available. Some workarounds reduce the likelihood of incidents.
Workload: A specific application, service, or capability that can run in the cloud or on-premises (such as virtual machines, databases, or containers). See course-notes/cloud/google-cloud-digital-leader/modernize-infrastructure-and-applications-with-google-cloud.
Zone (Cloud): An isolated deployment area within a cloud region designed to host resources, providing physical redundancy and high availability. See course-notes/cloud/google-cloud-digital-leader/digital-transformation-with-google-cloud.