Intermediate everything in DevOps & Cloud, page 23
Cloud Governance: Rules for Your Cloud Kingdom
Cloud governance is like city planning for your cloud, setting automated rules to prevent chaos. It's used to control costs by blocking expensive VMs and enforce security with required settings.
What is Cloud Native Architecture?
Cloud Native refers to a collection of practices and open-source projects, like Kubernetes, governed by the CNCF. It's used by enterprises to build modern products and services.
Cloud Landing Zone: A Blueprint for Cloud Environments
A Cloud Landing Zone is a pre-configured, secure foundation for your cloud applications, like a city grid with utilities ready for new buildings. It provides shared services like networking and identity, ensuring consistency for large organizations.
Cloud Center of Excellence (CCoE): Your Internal Cloud Guides
A Cloud Center of Excellence (CCoE) is your company's internal cloud consulting team, centralizing expertise to guide adoption. It creates reusable patterns, manages costs, and ensures security, helping large orgs scale.
ML Inference Endpoint: The API for Your Model
An ML inference endpoint is the stable API URL your application calls to get predictions. It separates the public URL from the underlying model, letting you swap models without changing client code.
Hyperparameter Tuning for LLM Inference
Control an LLM's creativity versus predictability by tweaking its inference parameters. This is crucial for tasks like generating structured JSON versus creative text. The footgun is changing parameters without a clear goal, leading to chaotic output.
Feature Store: The Single Source of Truth for ML
A feature store is the single source of truth for ML models, acting as a central kitchen for prepped ingredients (features). It's used to ensure the same feature logic is applied in both training and real-time inference, preventing model drift.
MLOps: Applying DevOps to Machine Learning
MLOps applies DevOps principles to automate the machine learning lifecycle, creating an assembly line for models. It's for moving from notebooks to production systems that retrain automatically.
OLAP Cube: Pre-Aggregating Data for Fast Analysis
An OLAP cube is like a Rubik's Cube for your data, pre-calculating answers to complex business questions. It powers BI tools, letting you 'slice and dice' sales data by region and time for fast reports. The footgun: data is typically stale, not real-time.
Lambda Architecture: Batch and Stream for Big Data
Lambda Architecture splits data into two paths: a slow, comprehensive batch layer and a fast, real-time stream layer. It's used in big data systems needing both historical accuracy and live views.
Apache Airflow: Code-Defined Data Pipelines
Airflow lets you define, schedule, and monitor complex data workflows as code, replacing brittle cron jobs. It's used for ETL jobs or ML training pipelines. The footgun is treating it as a data processing engine; it's an orchestrator, not the worker.
Dimensional Modeling: Facts vs. Dimensions
Dimensional modeling organizes data like a story: 'facts' are what happened (sales numbers) and 'dimensions' are the who, what, and where (customer, product). It's the foundation for data warehouses, turning raw data into analyzable BI reports.
Start/Stop Automation: Pay Only For What You Use
Start/stop automation is like putting your cloud resources on a timer to save money. It's essential for non-production environments like dev and staging used only during business hours.
Showback vs. Chargeback: Who Pays the Cloud Bill?
Showback shows teams their cloud costs for visibility; Chargeback makes them pay for it by moving costs to their budget. This helps control cloud spend by making engineers cost-aware. The footgun is treating Chargeback as inherently more mature than Showback.
FinOps Framework: Aligning Cloud Cost with Business Value
FinOps treats cloud spend as a business metric, not just an IT cost. It provides a shared framework for engineering, finance, and business to collaborate on data-driven spending decisions.
AWS Savings Plans: A Bulk Discount for Compute
Think of Savings Plans as a bulk discount for AWS compute. You commit to a consistent hourly spend for 1-3 years to get a lower rate on EC2, Fargate, and Lambda. The footgun is over-committing; you pay for your commitment even if you don't use.
Cloud Rightsizing: Stop Overpaying for VMs
Rightsizing stops you from overpaying for idle cloud capacity. It involves analyzing CPU and memory usage to shrink over-provisioned VMs. Always collaborate with application owners before making changes.
Cloud Cost Management: Taming Your Bill
Treat cloud spend like a utility bill you can actively control, not a fixed cost. It's essential when your AWS, GCP, or Azure bill is growing unpredictably. The biggest footgun is treating cost management as a one-time cleanup instead of a continuous process.
SLIs & SLOs: Measuring What Matters for Service Reliability
SLIs are what you measure (e.g., latency); SLOs are the target you aim for (e.g., 99% success). They replace vague feelings about service health with concrete numbers. This is how SREs define and manage reliability.
The Fan-out Pattern: One Message, Many Receivers
The fan-out pattern uses a single message to trigger multiple parallel actions, like a press conference where one announcement reaches many reporters. Use it for events like a new user signup that triggers emails, analytics, and fraud checks.
We are hiring for this. Every open role lists the topics its interview covers, so you can prepare for the real thing rather than guessing.
See open roles