Cloud Providers & Platforms
The platforms you actually deploy on, from the hyperscalers to the developer and edge PaaS layer above them. Interviewers check you can pick a provider for a workload and explain what you pay for.
on this pageshowhide
explore
- AWS (has its own guide)442 questions
- IAM & Security69 questions
- Networking & CDN88 questions
- Compute78 questions
- Storage & Databases82 questions
- Messaging & Eventing65 questions
- Observability & Cost60 questions
- Azureempty
- Compute Servicesempty
- Storage & Databasesempty
- Networkingempty
- Identity & Accessempty
- AKS & Containersempty
- Azure Functionsempty
- Azure Service Busempty
- Google Cloudempty
- IAM and Securityempty
- Compute Engineempty
- Cloud Storageempty
- Networking and VPCempty
- GKE and Containersempty
- Alibaba Cloudempty
- Cloudflare296 questions
- Content Delivery Settings28 questions
- Domains & Traffic17 questions
- Edge Threat Defenses33 questions
- Edge Compute45 questions
- Edge Data Stores27 questions
- Frontend Hosting17 questions
- Durable Objects17 questions
- Background Processing16 questions
- Develop & Deploy23 questions
- SSL/TLS Settings11 questions
- Zero Trust Services12 questions
- AI Products23 questions
- Media Delivery17 questions
- Automation & Mail Services10 questions
- Vercelempty
- Netlifyempty
- Renderempty
- Herokuempty
- Firebase36 questions
- Firestore & Realtime DB6 questions
- Firebase Authentication6 questions
- Security Rules6 questions
- Cloud Functions6 questions
- Storage & Hosting6 questions
- Cloud Messaging (FCM)6 questions
- Railwayempty
- Cloud Platform Concepts272 questions
- Provider Responsibility Model22 questions
- Global Footprint & Placement25 questions
- Tenancy Layout21 questions
- Platform Access Model32 questions
- Renting Compute22 questions
- Platform Storage Services23 questions
- Provider Network Model33 questions
- Reliability & Failure Domains25 questions
- Billing & Charge Model29 questions
- Platform Operations22 questions
- Exit & Switching Costs18 questions
- AI & Data Scientistrole
- AI Engineerrole
- AWStool
- Android Developerrole
- Cloudflaretool
- Cyber Security Expertrole
- Data Engineerrole
- DevOps / SRE Engineerrole
- DevSecOps Engineerrole
- Forward Deployed Engineerrole
- MLOps Engineerrole
- Machine Learning Engineerrole
- Network Engineerrole
- PostgreSQL DBArole
- SOC Analystrole
- Server-Side Game Developerrole
- Software Architectrole
questions
1 046 · 4 sectionsWhat is EC2 user data, at what point in an instance's life does the script run, and as which OS user?
basics
~20 sUser data is a text blob you attach to an EC2 instance that cloud-init fetches from the instance metadata service at first boot. A script starting with a shebang runs as root, once per instance — not on every reboot.
In an EC2 Auto Scaling group, what do the minimum, maximum and desired capacity settings each control, and what happens if you manually terminate one of the group's instances?
basics
~20 sDesired capacity is the instance count the group tries to run right now; minimum and maximum are the floor and ceiling that scaling can never cross. Manually terminating an instance leaves the group below desired, so it launches a replacement.
Decode the EC2 instance type name m7g.2xlarge: what does each part of the name tell you, and how many vCPUs and how much memory does that instance have?
basics
~20 sm7g.2xlarge splits into four parts: family m (general purpose, roughly 4 GiB of memory per vCPU), generation 7, processor letter g (AWS Graviton, arm64), and size 2xlarge — 8 vCPUs and 32 GiB of memory.
In EC2, what do you get and what do you give up by running an instance as Spot instead of On-Demand, and how is the Spot price actually determined?
basics
~20 sSpot Instances run on EC2 capacity AWS currently has idle, at a steep discount off On-Demand, but AWS can reclaim them at any moment with a two-minute warning. Spot prices move gradually with supply and demand; there is no bidding auction.
In Amazon ECS, explain what a task definition, a task, and a service each are, and when you would use the RunTask API directly instead of putting the workload behind a service.
basics
~20 sAn ECS task definition is the immutable blueprint: image, CPU and memory, IAM roles, logging. A task is one running copy of it. A service keeps a desired count of tasks alive and replaces failures; RunTask launches one-off jobs.
In Cloudflare's Agents SDK, how does routeAgentRequest() reach an Agent instance, and why is the instance name the agent's identity?
basics
~20 srouteAgentRequest() parses /agents/{class-in-kebab-case}/{instance-name} and forwards the request to the Durable Object for that name. The same class and name always reach the same agent and its private SQLite state, so the name is its identity.
In Workers AI, how does a Worker call a model through the AI binding with env.AI.run(), and what comes back?
basics
~20 sA Worker declares an ai binding in its Wrangler file and calls env.AI.run(model, inputs) with an @cf/ catalogue id; the promise resolves to that task's output, such as { response } for text generation or { shape, data } for embeddings.
In Cloudflare Queues, how does a producer Worker send an order event, and how does a consumer Worker receive it?
basics
~20 sA producer Worker binds the queue under queues.producers and calls env.ORDERS.send(body) or sendBatch(); a consumer Worker listed under queues.consumers exports a queue(batch, env, ctx) handler and reads batch.messages, each with id, timestamp, attempts and body.
In a Cloudflare Worker, how do you reuse an expensive JSON response with caches.default, cache.match() and cache.put()?
basics
~10 sBuild a GET Request as the key and return caches.default.match(key) when it hits; on a miss, build the response with a Cache-Control lifetime, store response.clone() with cache.put() inside ctx.waitUntil, and return the original.
On a site proxied by Cloudflare, which files does the edge cache by default, and why do HTML pages show cf-cache-status DYNAMIC?
basics
~20 sCloudflare caches by file extension: images, CSS, JavaScript, fonts, media, documents and archives, plus robots.txt. HTML, JSON and extensionless paths are not on the list, so those requests go to the origin without a cache lookup and show DYNAMIC.
In Firebase Cloud Messaging for Android, what is the difference between a notification message and a data message?
basics
~20 sA notification message carries display fields (title, body, image) that the FCM SDK shows itself while the app is in the background; a data message carries only custom string pairs, handed to onMessageReceived for your code to act on.
In Cloud Firestore, how do documents, collections and subcollections fit together, and what happens to a subcollection when its parent document is deleted?
basics
~20 sFirestore keeps fields in documents of up to 1 MiB, documents in collections, and lets a document own subcollections, so paths alternate collection and document. Deleting a document leaves its subcollections in place; they must be deleted separately.
In Firebase Security Rules for Firestore, how do you let each student in a classroom app read only their own grades?
basics
~10 sMatch the grades path with a wildcard for the student ID and allow reads only when request.auth.uid equals that wildcard. Firestore denies anything no rule allows, and each subcollection needs its own match block.
With the Cloud Storage for Firebase Android SDK, how do putBytes, putStream and putFile differ, and how do you get a download URL afterwards?
basics
~20 sAll three upload to a StorageReference and return an UploadTask: putBytes needs the whole byte array in memory, putStream reads an InputStream, putFile reads a file or content Uri. Ask the reference for getDownloadUrl() once the task succeeds.
In Firebase Authentication, how long does an ID token live, what ends its refresh token, and when should an Android app force a refresh?
basics
~20 sA Firebase ID token lasts one hour; the long-lived refresh token mints new ones until the user is deleted, disabled, changes password or email, or is revoked. Call getIdToken(true) only when server-side state such as custom claims changed.
A nightly cleanup job runs about twenty minutes - why does an event-driven runtime rule itself out, and which tier fits?
basics
~20 sAn event-driven runtime caps how long one invocation may run, so a twenty-minute job is stopped part-way. Put it on a tier that keeps a process alive: a managed container platform running it on a schedule, or a machine that stays up.
You are moving a measured search service onto rented machines — which numbers set the machine size, and which do you ignore?
basics
~20 sSize from the service's own observed processor, memory, disk and network use over a full demand cycle, taken at a high percentile with headroom added. The specification of the server being replaced records a purchase decision, not the workload, and is ignored.
What does a label attached to a cloud resource actually change about the bill, and what does it not change?
basics
~20 sA label copies a key-value pair onto the charges a resource generates, so the provider's detailed cost report can be grouped by team, environment or service. It changes reporting only: the rate, the metered quantity and the total are unaffected.
A monthly cloud budget threshold is crossed at noon — what does crossing it actually do, and what does it not do?
basics
~20 sA budget threshold is a notification rule, not a spending cap. Crossing it emits a message about money already spent. It does not pause running resources, block new ones, or cancel anything, and the figure it fired on lags the usage.
On a cloud transfer bill, a service uploads large video masters and ships finished renditions out to viewers — why is only one direction charged?
basics
~20 sTransfer is metered by direction. Bytes arriving from outside are normally unmetered; bytes leaving toward the internet are priced per gigabyte after a small monthly allowance. So delivery to viewers is the charge, not the upload of masters.