Deployment Glossary
Browse 96 deployment terms defined in plain English, from the cultural dictionary of computing.
96 Deployment Terms
- AI Deploy
- To release or put an AI model or AI-powered feature into an environment where it can be used for testing or production workloads. AI deploy decisions usually...
- AI Deployment
- The operational process and resulting setup for serving an AI model or AI feature in staging or production. AI deployment includes packaging, routing, scaling,...
- AI Edge
- AI inference or processing performed close to where data is generated or consumed, such as on devices, local gateways, or edge servers rather than only in...
- AOT Compilation
- Ahead-of-Time compilation — translating high-level code to native machine code before deployment, producing a standalone binary. Eliminates cold-start latency,...
- Application Container
- A container image or runtime unit that packages an application with its dependencies and executes it in an isolated environment. Application containers...
- blue-green deployment
- A deployment strategy that maintains two identical production environments: 'blue' (current live) and 'green' (new version). Traffic is switched from blue to...
- Blue-Green Deployment
- Blue-green deployment is a release strategy that maintains two identical production environments, labeled blue and green. At any time, one environment serves...
- Blue/Green Deployment
- A deployment strategy using two identical production environments (blue and green). The current version runs on blue; the new version is deployed to green and...
- Canary Analysis
- Automated statistical comparison between a canary deployment (new version, small traffic percentage) and the baseline (current version) to detect regressions...
- Canary Deployment
- A deployment strategy that gradually rolls out changes to a small percentage of users before deploying to the entire fleet. Named after canaries in coal mines...
- canary release
- A deployment strategy where a new version is rolled out to a small subset of users or servers first, then gradually expanded if metrics look healthy. Named...
- Cloud Container
- A containerized workload running in a cloud-managed environment, often on orchestration platforms or serverless container services. Cloud containers package...
- Cloud Deployment
- The act of releasing an application, service, or infrastructure change into a cloud environment. Cloud deployments may involve containers, serverless...
- Config Map
- A named collection of configuration key-value pairs, especially in Kubernetes, used to inject non-secret settings into applications and containers at runtime....
- Container Config
- The runtime settings, environment variables, mounts, ports, resource limits, and entrypoint details that define how a container runs. Container config is...
- Coolify
- An open-source, self-hosted alternative to platforms like Vercel, Netlify, and Heroku. Coolify lets developers deploy applications, databases, and services on...
- dark launch
- A deployment technique where new code is released to production and executed in the background without exposing results to users. Dark launches allow teams to...
- Edge AI
- Running AI models directly on local devices (phones, cameras, IoT) rather than in the cloud, reducing latency and keeping data private. The reason your phone...
- Feature Flag
- A feature flag (also called a feature toggle) is a configuration mechanism that enables or disables application functionality at runtime without deploying new...
- Feature Toggle
- A runtime configuration switch that enables or disables a feature without deploying new code. Categories: release toggles (gradual rollout), experiment toggles...
- Fly.io
- A platform for deploying full-stack applications and databases close to users worldwide using lightweight virtual machines (Firecracker microVMs). Unlike...
- GitOps
- An operational framework where the desired state of infrastructure is declared in a Git repository, and automated processes continuously reconcile the actual...
- GitOps Pull Model
- A GitOps implementation where an agent running inside the cluster (ArgoCD, Flux) continuously polls a Git repository and reconciles the cluster state to match...
- guardrails
- Safety mechanisms and filters applied to AI systems to prevent harmful, off-topic, or policy-violating outputs. Guardrails can be implemented at the prompt...
- Helm Chart
- A package format for Kubernetes applications containing templated YAML manifests, default configuration values, and metadata. Helm charts enable versioned,...
- Hotfix
- Hotfix is an urgent software patch deployed directly to production to address a critical bug, security vulnerability, or outage without going through the full...
- Immutable Deploy
- A deployment model where servers are never updated in place — instead, a new machine image (AMI, Docker image) is built with the new code and replaces the old...
- immutable infrastructure
- An infrastructure management approach where servers are never modified after deployment — any change requires building a new server image and replacing the old...
- Kustomize
- A Kubernetes configuration customization tool that layers patches and overlays onto base manifests.
- LLMOps
- The operational discipline around deploying, monitoring, evaluating, and maintaining large-language-model systems in production.
- Mixed Precision Training
- Mixed Precision Training is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for...
- MLflow
- MLflow is an MLOps platform for experiment tracking, model packaging, and lifecycle management. It is commonly used for recording runs, registering models, and...
- Model Architecture
- Model Architecture is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Card Detail
- Model Card Detail is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Compression
- Model Compression is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Deployment
- The process and resulting setup for releasing a model into an environment where it can serve real workloads. Model deployment includes packaging, routing,...
- Model Distillation
- Model Distillation is a training approach where a smaller student model learns from a larger teacher model. It is commonly used for compressing models while...
- Model Ensemble
- Model Ensemble is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Evaluation
- Model Evaluation is an evaluation concept used to measure model quality, robustness, or efficiency. It is commonly used for comparing systems before release or...
- Model Explainability
- Model Explainability is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Governance
- Model Governance is a safety or governance control for limiting harmful, noncompliant, or insecure model behavior. It is commonly used for production AI...
- Model Hub
- Model Hub is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Inference
- Model Inference is the phase where a trained model processes new inputs to produce predictions or generations. It is commonly used for production APIs, batch...
- Model Merging
- Model Merging is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Parallelism
- Model Parallelism is a systems strategy for splitting model computation or parameters across devices. It is commonly used for training and serving workloads...
- Model Poisoning
- Model Poisoning is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Pruning
- Model Pruning is a compression technique that removes parameters, channels, or structures judged less important. It is commonly used for shrinking models for...
- Model Quantization
- Model Quantization is a technique that reduces numerical precision so models use less memory and compute. It is commonly used for deploying large models on...
- Model Registry
- Model Registry is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- model serving
- The infrastructure and process of deploying trained machine learning models to production so they can receive input data and return predictions in real time or...
- Model Serving Detail
- Model Serving Detail is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Sharding
- Model Sharding is a systems strategy for splitting model computation or parameters across devices. It is commonly used for training and serving workloads that...
- Model Stacking
- Model Stacking is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Training
- Model Training is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- Model Versioning
- Model Versioning is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Model Zoo
- Model Zoo is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Multi-GPU Training
- Multi-GPU Training is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- Non-Autoregressive Model
- Non-Autoregressive Model is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building...
- ONNX Runtime
- A runtime for executing models in the ONNX format across different hardware backends with an emphasis on portability and inference performance.
- Open Source Model
- Open Source Model is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Open Weight Model
- Open Weight Model is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Port Binding
- The practice of a service binding to a network port to listen for incoming requests, making itself available as a self-contained, externally accessible...
- Post-Deploy Hook
- A script or automation step that executes automatically after a deployment completes, typically used for tasks like cache invalidation, database migrations,...
- Post-Training
- Post-Training is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- Pre-Deploy Hook
- A script or automation step executed before a deployment begins, commonly used to run final test suites, create database backups, enable maintenance pages, or...
- Pre-Training
- Pre-Training is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- Pretraining Data
- Pretraining Data is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- Probabilistic Model
- Probabilistic Model is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Process Manager
- A daemon or tool that starts, monitors, and restarts application processes, ensuring they stay running and are properly managed. Examples include systemd, PM2...
- Production Build
- A compiled and optimized version of an application intended for end users, with minification, tree-shaking, dead code elimination, source map separation, and...
- Production Environment
- The live infrastructure — servers, databases, CDNs, and services — where the application serves real users and processes real data, as distinguished from...
- Quantization
- Reducing the precision of a model's numerical weights (e.g., from 32-bit to 4-bit) to decrease memory usage and increase inference speed, with minimal quality...
- Quantization Aware Training
- Quantization Aware Training is a technique that reduces numerical precision so models use less memory and compute. It is commonly used for deploying large...
- Railway
- A modern application hosting platform that simplifies deploying backend services, databases, and full-stack applications. Railway provides a visual dashboard...
- Release Build
- A compiled version of software with optimizations enabled, debug symbols stripped, assertions removed, and minification or tree-shaking applied — producing...
- Release Channel
- A named distribution track (such as stable, beta, canary, or nightly) that determines which version of software a user receives, allowing teams to ship...
- Release Security
- The protections applied to the software release process, including build integrity, artifact signing, approval controls, provenance, and safe deployment...
- Rollback
- Rollback is the process of reverting a system, database, or deployment to a previous known-good state after a failed update, bug, or incident. In database...
- Rolling Update
- A deployment strategy that gradually replaces old instances with new ones, maintaining availability throughout. Typically configured with maxSurge (extra...
- Runtime Environment
- The complete set of software, libraries, configuration, and system resources available to an application during execution — including the runtime itself,...
- Scalar Quantization
- Scalar Quantization is a technique that reduces numerical precision so models use less memory and compute. It is commonly used for deploying large models on...
- Ship It Squirrel
- A meme originating from GitHub's Campfire chat featuring a squirrel wearing a pilot's helmet, posted in code reviews to signal 'looks good, let's deploy.' The...
- Smoke Test
- Smoke Test is a preliminary, broad-but-shallow test run performed to verify that the most critical functions of a software system work before committing to...
- Staged Rollout
- A deployment strategy that gradually increases the percentage of users or servers receiving a new version — typically progressing through stages like 1%, 5%,...
- Staging Environment
- A pre-production environment that mirrors the production infrastructure as closely as possible — same services, configuration, and (ideally) anonymized...
- State Space Model
- State Space Model is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Static Binary
- An executable that has all of its library dependencies compiled and linked directly into the binary at build time, requiring no shared libraries on the target...
- Teacher Model
- Teacher Model is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
- Training Compute
- Training Compute is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- Training Data Detail
- Training Data Detail is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- Training Loop
- Training Loop is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- Training Run
- Training Run is a stage of model optimization where weights or behaviors are adjusted from data or feedback. It is commonly used for foundation-model...
- Upgrade
- The process of replacing a software component — such as a library, framework, runtime, or operating system — with a newer version, often involving dependency...
- Vector Quantization
- Vector Quantization is a technique that reduces numerical precision so models use less memory and compute. It is commonly used for deploying large models on...
- Vercel
- A cloud platform for frontend developers that specializes in deploying web applications with zero configuration. Founded by Guillermo Rauch and originally...
- World Model
- World Model is a model component or design choice that shapes how information flows through a learned system. It is commonly used for building neural...
Related Topics
- Mlops (43 terms in common)
- Models (43 terms in common)
- Devops (10 terms in common)
- Ai (5 terms in common)
- Kubernetes (5 terms in common)
- Configuration (4 terms in common)
- Operations (4 terms in common)
- Infrastructure (4 terms in common)
- Containers (4 terms in common)
- Ci/Cd (3 terms in common)
- Testing (3 terms in common)
- Hosting (3 terms in common)
- Safety (3 terms in common)
- Reliability (3 terms in common)
- Cloud (3 terms in common)
- Pattern (2 terms in common)
- Optimization (2 terms in common)
- Strategy (2 terms in common)
- Monitoring (2 terms in common)
- Automation (2 terms in common)