クラウドコンピューティング英語：必須語彙とフレーズ

"infrastructure as a service"

A cloud model in which a provider delivers virtualised computing resources such as servers, storage, and networking over the internet

"The startup chose infrastructure as a service to avoid purchasing physical hardware during its early growth phase."

"platform as a service"

A cloud model that provides a managed environment for developers to build, deploy, and run applications without managing the underlying infrastructure

"The engineering team used platform as a service so developers could focus on code rather than server configuration."

"software as a service"

A cloud delivery model in which a fully managed application is accessed via a web browser or API rather than installed locally

"The company replaced its on-premise email server with a software as a service solution to reduce maintenance overhead."

"multi-cloud strategy"

An approach in which an organisation deliberately uses services from two or more cloud providers to avoid vendor lock-in and improve resilience

"Adopting a multi-cloud strategy allowed the bank to run critical workloads on AWS while keeping analytics on Google Cloud."

"hybrid cloud"

An environment that combines on-premise data centres with public cloud services, allowing data and applications to move between them

"The hospital implemented a hybrid cloud so patient records stayed on-premise while analytics workloads ran in the public cloud."

"cloud-native"

An approach to building applications specifically designed to exploit cloud computing capabilities such as elasticity, microservices, and managed services

"The new platform was cloud-native from day one, built on containers and a fully managed database service."

"elasticity"

The ability of a cloud system to automatically scale computing resources up or down in response to changing demand

"Thanks to elasticity, the e-commerce site handled ten times its normal traffic during the Black Friday sale without any intervention."

"shared responsibility model"

A security framework defining which aspects of security the cloud provider manages and which remain the customer's responsibility

"The security audit revealed the team had misunderstood the shared responsibility model and left database encryption to the customer side unconfigured."

"container image"

A lightweight, standalone, executable package that includes everything needed to run a piece of software, including code, runtime, and dependencies

"The CI pipeline builds a fresh container image on every pull request merge and pushes it to the registry."

"Kubernetes cluster"

A set of worker machines called nodes that run containerised applications managed by Kubernetes, with a control plane coordinating the workloads

"The SRE team migrated all production workloads from virtual machines to a managed Kubernetes cluster to improve resource utilisation."

"pod"

The smallest deployable unit in Kubernetes, representing one or more containers that share network and storage resources

"Each pod in the payment service runs a single container and is automatically restarted if it becomes unhealthy."

"horizontal pod autoscaler"

A Kubernetes controller that automatically adjusts the number of pod replicas based on observed CPU utilisation or custom metrics

"The horizontal pod autoscaler scaled the API service from 3 to 15 replicas during the morning traffic spike."

"service mesh"

An infrastructure layer that manages service-to-service communication in a microservices architecture, handling load balancing, encryption, and observability

"After adopting a service mesh, the team gained end-to-end encryption and detailed latency metrics between every microservice."

"Helm chart"

A package of pre-configured Kubernetes resource definitions that can be deployed and managed as a single unit

"The database team published a Helm chart so other squads could deploy a consistent PostgreSQL instance in minutes."

"container registry"

A repository for storing, versioning, and distributing container images used in deployment pipelines

"All production container images are scanned for vulnerabilities before being pushed to the private container registry."

"sidecar container"

A secondary container running alongside the main application container in the same pod to provide supporting functions such as logging or proxying

"A sidecar container running Envoy handles all outbound TLS termination so the main application does not need to manage certificates."

"function as a service"

A serverless execution model in which the cloud provider runs individual functions in response to events, billing only for actual execution time

"The team replaced a 24/7 polling service with function as a service, cutting monthly compute costs by 80 per cent."

"cold start"

The latency introduced when a serverless function is invoked for the first time or after a period of inactivity, requiring the runtime environment to be initialised

"High cold start latency in the authentication function was causing timeout errors for users who had been idle for several minutes."

"event-driven architecture"

A design pattern in which services communicate by producing and consuming events, enabling loose coupling and independent scaling

"Switching to event-driven architecture meant the order service could scale independently of the inventory service during peak demand."

"dead letter queue"

A message queue that holds messages that could not be successfully processed after a defined number of retries, preventing data loss

"The team configured a dead letter queue so that any failed payment notification could be investigated and reprocessed manually."

"trigger"

An event or condition that causes a serverless function or automated workflow to execute

"An S3 upload trigger fires the image-resizing function every time a user uploads a profile photo."

"provisioned concurrency"

A configuration that pre-warms a specified number of serverless function instances to eliminate cold start latency for latency-sensitive workloads

"The checkout service used provisioned concurrency to guarantee sub-100-millisecond response times during the product launch."

"idempotency"

The property of an operation that produces the same result regardless of how many times it is executed, critical for reliable serverless and distributed systems

"Every payment processing function is designed for idempotency so duplicate event deliveries never result in double charges."

"fan-out pattern"

An architectural pattern in which a single event triggers multiple downstream processes simultaneously, enabling parallel workload distribution

"The fan-out pattern allowed a single order confirmation event to simultaneously trigger shipping, invoicing, and loyalty point calculations."

"infrastructure as code"

The practice of managing and provisioning cloud infrastructure through machine-readable configuration files rather than manual processes

"After adopting infrastructure as code with Terraform, the team could spin up a complete staging environment in under ten minutes."

"pipeline as code"

The practice of defining CI/CD pipeline configuration in version-controlled files, ensuring the build process is reproducible and auditable

"Pipeline as code meant every change to the deployment process went through a code review before being merged."

"blue-green deployment"

A release strategy using two identical production environments (blue and green) where traffic is switched to the new version after validation, enabling instant rollback

"The blue-green deployment strategy allowed the team to release a major API update with zero downtime and a one-click rollback option."

"canary release"

A deployment technique in which a new version is rolled out to a small subset of users before a full release, allowing issues to be detected with limited impact

"The engineering manager approved a canary release sending 5 per cent of traffic to the new recommendation engine before the full rollout."

"GitOps"

An operational model in which the desired state of cloud infrastructure and application deployments is declared in Git, with automated agents ensuring the actual state matches

"Adopting GitOps meant every infrastructure change was traceable to a Git commit, dramatically simplifying incident post-mortems."

"shift left"

A DevOps principle of moving testing, security checks, and quality gates earlier in the development lifecycle to catch issues before they reach production

"Shift left practices reduced the number of critical security findings in production by running SAST tools on every pull request."

"mean time to recovery"

A metric measuring the average time required to restore a service after an incident, used to evaluate operational resilience

"Investing in automated rollback scripts cut the team's mean time to recovery from 45 minutes to under five."

"feature flag"

A technique that allows a feature to be enabled or disabled at runtime without deploying new code, enabling gradual rollouts and instant kill switches

"The product manager used a feature flag to enable the new checkout flow for beta users without affecting the rest of the customer base."

"identity and access management"

A framework of policies and technologies controlling which users and systems are permitted to access which cloud resources and under what conditions

"The security review found that identity and access management policies had been misconfigured, granting developer accounts write access to production buckets."

"principle of least privilege"

A security best practice in which every user, service, and process is granted only the minimum permissions required to perform its function

"Applying the principle of least privilege to all service accounts reduced the blast radius when one account was compromised."

"zero trust architecture"

A security model that assumes no user or system is inherently trustworthy and requires continuous verification before granting access to any resource

"Migrating to zero trust architecture eliminated the perimeter-based assumption that anything inside the corporate network was safe."

"encryption at rest"

The protection of stored data by encrypting it on disk so it cannot be read without the correct decryption key

"All S3 buckets containing customer data are configured with encryption at rest using AWS KMS-managed keys."

"encryption in transit"

The protection of data as it travels between systems by encrypting the communication channel, typically using TLS

"The compliance audit confirmed that encryption in transit was enforced on every API endpoint handling personally identifiable information."

"cloud security posture management"

Tooling and processes that continuously monitor a cloud environment for misconfigurations, compliance violations, and security risks

"Cloud security posture management flagged an exposed storage bucket within minutes of it being accidentally made public."

"secrets management"

The secure storage, rotation, and distribution of sensitive credentials such as API keys, database passwords, and certificates in automated environments

"After the incident, the team adopted a dedicated secrets management service to prevent credentials from being hard-coded in application code."

"service control policy"

An organisational governance mechanism in AWS that sets maximum permission boundaries across multiple accounts within an organisation

"A service control policy prevented any team from disabling CloudTrail logging, even with administrator privileges."

"virtual private cloud"

An isolated section of a public cloud provider's network in which resources run inside a logically separated, user-defined network environment

"All production workloads run inside a virtual private cloud with private subnets to prevent direct internet access to database servers."

"content delivery network"

A globally distributed network of proxy servers that caches static and dynamic content close to end users to reduce latency and origin server load

"Serving static assets through a content delivery network reduced average page load time for Asian users from 3.2 seconds to under 400 milliseconds."

"availability zone"

A physically separate data centre within a cloud region, designed so that a failure in one zone does not affect others, enabling high availability deployments

"Deploying the application across three availability zones ensured it remained online even when one zone experienced a power outage."

"egress cost"

The charge imposed by a cloud provider for data transferred out of its network to the internet or to another provider, often the dominant cost in data-heavy architectures

"The architecture review identified egress cost as the fastest-growing line item and recommended caching responses at the edge to reduce outbound data transfer."

"distributed tracing"

An observability technique that tracks a single request as it propagates through multiple microservices, recording timing and metadata at each step

"Distributed tracing revealed that 80 per cent of checkout latency was caused by a single downstream inventory service call."

"service level objective"

A measurable target for a specific aspect of service quality, such as availability or latency, that defines the expected performance level for a system

"The team set a service level objective of 99.9 per cent availability and 200-millisecond p99 latency for the payments API."

"load balancer"

A component that distributes incoming network traffic across multiple backend instances to maximise throughput, minimise latency, and prevent overload

"The application load balancer routes requests to the healthiest instance in each availability zone and automatically removes unhealthy targets."

"peering connection"

A private networking link between two virtual private clouds that allows instances to communicate without traffic traversing the public internet

"A peering connection between the data platform VPC and the analytics VPC eliminated the need to route sensitive data over the public internet."