Cloud + cv0-04 Section 1

Cloud computing

Using someone else's computers and storage over the internet instead of buying and managing your own. Think of it like using the power company instead of building your own generator.

On-demand self-service

You can get computer resources (like storage or power) whenever you want, automatically, without asking a human. Like getting snacks from a vending machine.

Broad network access

You can use the cloud from almost any device with internet (phone, laptop, tablet). Like watching TV on different screens.

Resource pooling

The cloud company pools all its computers and storage together and shares them out safely to many customers (multi-tenant). Like an apartment building sharing one big swimming pool.

Rapid elasticity

You can quickly get more computer power when you need it, and quickly give it back when you don't. Like stretching a rubber band bigger or smaller. Can be automatic or manual. Grows or shrinks with your needs.

Measured service

You pay for what you use. The cloud company measures your usage (like an electricity meter) and sends you a bill.

IaaS (Infrastructure as a Service)

Renting the basic building blocks - computer power (server/VMs), storage, and networking.

PaaS (Platform as a Service)

Renting the building block plus the tools and environment to build and run your apps (like the OS and database). You manage your apps and data.

SaaS (Software as a service)

Renting a ready-to-use application over the internet (like Gmail, Office 365, Netflix). You just use it; the cloud company manages everything else.

FaaS (Function as a service)/ serverless

Running small pieces of code in response to events, without managing any servers. The cloud automatically runs the code when needed and stops it afterward.

XaaS (anything as a service)

A general term meaning turning any IT function into a service delivered over the cloud

MSP (Managed service provider)

A separate company you can hire to help manage your cloud stuff (design, setup, security).

Shared responsibility model

You and the cloud company share security jobs. They secure the building and the base hardware; you secure your data, who accesses it, and how you configure things.

Microservices

An architectural style that breaks an application into small, independent services, each running in its own process and communication over APIs.

Sharding

Splitting a large dataset horizontally across multiple database instances based on a shard key (e.g., userID) to distribute load and storage

Rightsizing

Selecting the optimal instance type (vCPU, memory, storage) to match actual workload needs, avoiding over-provisioning (waste) or under-provisioning (performance issues).

IoT platforms

Managed services for ingesting, processing, and analyzing data from Internet of Things devices, often including device management and rules engines

Edge computing

Running compute and storage closer to data sources (e.g., on-premises gateways or regional edge nodes) to reduce latency and bandwidth use.

Read replicas

Copies of a primary database that asynchronously receive updates, used to offload read traffic and improve scalability

On-premise

Everything runs on computers at your company

Application gateway

A type of load balancer or reverse proxy service that manages and directs client requests to appropriate backend web applications or services, often operating at application layer (layer 7) and providing features like SSL termination, routing based on URL paths, and web application firewall (WAF) capabilities

Relational database

A database that organizes data into structured tables with predefined schemas, rows, and columns, enforcing relationships between tables (e.g., SQL databases like MySQL, PostgreSQL).

Non-relational database

A database that stores data in formats other than structured tables, such as key-value pairs, documents, graphs, or wide-columns (e.g., NoSQL databases like MongoDB, Cassandra, DynamoDB), often prioritizing flexibility and scalability.

Fan-out

An architectural pattern where a single message or event triggers multiple actions or processes to occur in parallel, often used in event-driven or microservices architectures

Container networking (Port mapping)

The process of mapping a port on the host machine to a port on a running container, allowing external traffic to reach the application inside the container by specifying the port numbers for communication purposes.

Image registry

A centralized repository for storing and distributing container images (like docker images)

Hardware pass-through

Allowing a virtual machine direct access to a physical piece of hardware (like a GPU or network card), bypassing the hypervisor for potentially higher performance

Overlay network

A virtual network built on top of an existing physical network, often used in virtualization and containerization to create isolated network segments

Host affinity

Rules that dictate whether certain virtual machines should run on the same physical host or be kept on separate hosts (anti-affinity), often for performance or high-availability reasons.

Reserved instances/resources

Paying for cloud resources (like VMs) upfront or for a long-term commitment (e.g., 1 or 3 years) to receive a significant discount compared to pay-as-you-go pricing.

Spot instances

Purchasing unused cloud compute capacity at potentially very steep discounts, with the understanding that the cloud provider can reclaim the capacity with short notice, making them suitable for fault-tolerant or non-urgent workloads

Resource metering

The process by which cloud providers measure the consumption of resources (CPU, RAM, storage, network egress) by a customer to calculate billing charges, aligning with the measured services characteristic

ML/AI Services (Examples)

Services like text recognition (OCR), text translation, visual recognition (image/video analysis), sentiment analysis, voice-to-text, text-to-voice, and generative AI (creating new content)

IoT components

Including sensors (collect data), Gateways (aggregate data and connect sensors to the network/cloud), and communication/transmission protocols (how devices talk, e.g., MQTT, CoAP).

Service discovery

Process by which applications and microservices automatically find the network locations (IP addresses and ports) of other services they need to communicate with within a distributed system, often using a central registry.

Managed

Cloud provider handles setup, patching, backups, and high availability (e.g., Amazon RDS)

Self-hosted

You install and maintain the database in a VM or container yourself, giving full control but more overhead

MTBF (Mean time between failures)

Average time before something breaks

MTTR (Mean time to repair)

How fast it gets fixed

RPO (Recovery point objective)

How much data loss is okay

RTO (Recovery time objective)

How fast it must come back online