CCNA Google Cloud Products and Services Questions

BigQuery is Google Cloud's serverless, highly scalable, SQL-based data warehouse. It automatically manages infrastructure and scales compute and storage independently, using a columnar storage format and a distributed query engine (Dremel) to run analytics on petabytes of data. With a pay-per-query pricing model and no clusters to provision, BigQuery is the definitive choice for a fully managed data warehouse on Google Cloud.

Unlike standard Compute Engine VMs, which benefit from live migration during infrastructure maintenance, preemptible VMs are not live-migrated. If the underlying host needs maintenance or Google reclaims capacity, the VM is immediately terminated within 30 seconds of the preemption notice. This means the VM's state is lost unless you have explicitly saved it elsewhere.

Cloud CDN caches content and helps mitigate DDoS by absorbing traffic.

Pub/Sub is a scalable messaging service for ingesting streaming data from IoT devices.

Cloud SQL is a fully managed relational database service that supports PostgreSQL, providing automated backups, point-in-time recovery, and high availability with synchronous replication across zones. It offers native PostgreSQL compatibility, so applications can migrate without any changes to their SQL or data access layer. With managed maintenance, scaling, and built-in security features like IAM integration and encryption at rest, Cloud SQL is the pragmatic default choice for most PostgreSQL migrations to Google Cloud.

Cloud Build is the fully managed CI/CD service that compiles your source code, runs tests, and builds the container image from a Dockerfile or build config. It executes the build steps defined in your pipeline, generating OCI-compliant images that are then pushed to a registry. This makes Cloud Build the correct core engine for the CI/CD pipeline.

Persistent Disk (SSD) is a durable, network-attached block storage service that provides consistent, low-latency performance and survives instance termination. For HPC scratch, it can be used when data must persist across restarts or preemption, allowing successful checkpointing and restart without data loss. Its performance scales with provisioned size, and it supports multiple per-instance volumes to meet aggregate throughput needs.

Preemptible VMs (and Spot VMs) are Compute Engine instances that can be terminated anywhere anytime due to excess capacity reuse, but they can be used for batch and fault-tolerant workloads. They are up to 60–80% cheaper than standard VMs, making them the ideal choice for a company running scheduled batch processing that can checkpoint and resume. To use them effectively, the application must handle unexpected termination gracefully and be restartable from saved state.

Cloud Run is a fully managed serverless container platform that automatically scales your container from zero instances up to a number sufficient to handle incoming HTTP requests, and then back down to zero when traffic disappears. You are billed only for the exact time spent processing a request, with a minimum granularity of 100 milliseconds, so when there is no request traffic there are zero compute charges. This gives the developer exactly what they asked for: a containerized application that scales to zero, eliminating infrastructure costs during idle periods.

Cloud Dataflow is a fully managed, unified stream and batch processing service built on Apache Beam. It provides exactly-once processing guarantees, automatic scaling, and powerful primitives for event-time processing, windowing, and triggers. With native Pub/Sub and BigQuery I/O connectors, Dataflow can ingest streaming data directly from Pub/Sub, apply complex transformations, and write results to BigQuery without additional glue code, making it the correct choice for this use case.

BigQuery is the correct choice because it is a serverless, highly scalable cloud data warehouse that supports standard ANSI SQL. It can query external datasets directly from Cloud Storage using external tables, where you define a table schema pointing to files (CSV, JSON, Parquet, Avro, ORC, etc.) without requiring an ETL pipeline. This makes it ideal for ad hoc SQL analysis on large datasets, since BigQuery automatically manages the underlying compute and parallelizes the query across the data, and you pay only for the data scanned. Additionally, BigQuery supports federated queries across other Google Cloud services, but querying Cloud Storage files is the most direct path for this scenario.

Cloud VPC (Virtual Private Cloud) is the correct service for creating and managing a private network in Google Cloud. It lets you define global or regional networks, allocate IP CIDR ranges, create subnetworks, set up firewall rules, dynamic routes (including BGP with Cloud Router), and connect on-premises environments via Cloud VPN or Interconnect. As a foundational networking primitive, Cloud VPC provides full control over routing, addressing, and isolation, directly matching the requirement to create and manage the network itself.

Cloud Dataproc is Google Cloud's managed Spark and Hadoop service, purpose-built to run workloads like Apache Spark directly. You can create a job-scoped cluster that automatically terminates as soon as the batch job finishes, so you only incur compute costs while the job is running, which is ideal for nightly analytics that do not need a persistent cluster. Dataproc also supports custom machine types, preemptible/spot workers, and integration with Cloud Storage, BigQuery, and Cloud Monitoring, making it the lowest-effort, cost-optimized choice.

Archive is the cheapest, designed for data accessed less than once a year, with retrieval times up to 24 hours.

Persistent Disk stores durable block-level data that exists independently of the virtual machine instance. When an instance is terminated or fails, the disk can be attached to a new instance in the same zone (or across zones with regional persistent disks), allowing the application to recover all state such as user sessions and database files. This independence is exactly what makes a stateful application resilient to instance-level failures.

You can apply an IAM condition to a Cloud Storage bucket policy to restrict access based on the requester's source IP address. For example, setting a condition like `request.sourceIp.startsWith("203.0.113.")` ensures only requests originating from your corporate IP range can read or write objects. This is a valid and direct method to enforce network-level access control for sensitive data in Cloud Storage, making it a correct alternative to CMEK.

Cloud Dataflow is the correct choice because it is a fully managed, serverless service for both stream and batch processing, built on the Apache Beam model. It provides native, optimized BigQuery I/O with exactly-once processing semantics, event-time windowing, and automatic scaling to handle unbounded clickstream data from Pub/Sub. Dataflow's built-in support for watermarks, triggers, and stateful transformations makes it ideal for running low-latency, continuous ETL pipelines that land directly in BigQuery.

Looker Studio is Google's free, self-service reporting and visualization tool that natively integrates with BigQuery as a first-class data source. You can connect directly to BigQuery datasets without exporting data, and it automatically handles query pagination and aggregation for fast interactive dashboards. Its cost model (free for creators) and native BigQuery connector make it the ideal choice for an analytics team that needs to build and share visualizations quickly.

Cloud Functions is the correct choice because it is Google Cloud's event-driven serverless compute service designed to respond directly to Cloud Storage events. Specifically, Cloud Functions natively subscribes to the `google.storage.object.finalize` event, which is emitted whenever an object is uploaded or overwritten in a bucket. This background function type requires no custom intermediary, as the event is delivered automatically and the function can immediately process the object, making it the most direct and efficient mechanism for triggering on object uploads.

Preemptible VMs offer a 60-80% cost reduction over standard VMs because they use Google's excess capacity, but they can be terminated at any time after a 30-second warning. For a batch job that processes large logs, you can design the job to be resumable by writing intermediate results to persistent storage, making this the most cost-effective choice without sacrificing completion. The transient nature of the instances is acceptable because the workload is fault-tolerant and can resume from checkpoints.

Cloud Interconnect provides direct, dedicated network connections between your on-premises network and Google Cloud, either via co-location facilities (Dedicated Interconnect) or through a service provider (Partner Interconnect). These private links offer consistent high bandwidth, low latency, and a guaranteed SLA, making them the correct choice for a hybrid cloud connection.

Vertex AI is Google Cloud's unified machine learning platform that integrates the entire ML workflow—from data preparation and feature engineering to model training, hyperparameter tuning, serving, and monitoring—under a single API and console. It consolidates AutoML, custom training, and MLOps tools so teams can manage models consistently. This fits the definition of a "unified platform" for machine learning.

Archive is the lowest-cost storage class in Google Cloud Storage, specifically designed for long-term backup and archival data that is accessed less than once per year. It offers the cheapest storage price, but retrieval times are typically hours (or even up to 365 days for some operations), and it imposes a 365-day minimum storage duration. For an organization storing archival data that is rarely accessed, Archive provides the optimal balance of cost efficiency and suitability, making it the correct choice.

Partner Interconnect establishes a connection between your on-premises network and Google's network via a supported third-party service provider, such as a colocation or network provider. It offers a service-level agreement (up to 99.99%) and can be provisioned at capacities from 50 Mbps to 10 Gbps depending on the partner, making it suitable for hybrid workloads. Unlike VPN, it does not traverse the public internet, providing more reliable, lower-latency connectivity.

AlloyDB is Google Cloud’s fully managed PostgreSQL-compatible database purpose-built for demanding transactional and analytical workloads, and it integrates AI-optimized features directly into the engine. It includes native support for vector embeddings and vector search (AlloyDB AI), plus a columnar engine that accelerates analytical queries and can speed up AI inference pipelines. For a migration from Cloud SQL for MySQL, AlloyDB provides the lowest-friction PostgreSQL-compatible path while adding the AI capabilities your new workload needs.

Cloud Interconnect provides dedicated, private network connectivity between your on-premises data center and Google Cloud, using either Dedicated Interconnect or Partner Interconnect. Because traffic is carried over Google's global network rather than the public internet, it offers consistently low latency, high throughput, and a contractually backed availability SLA (up to 99.99% depending on configuration). For a large enterprise migration that requires reliable performance and predictable network behavior, this is the only option that truly meets those requirements.

Cloud Armor is a global DDoS mitigation and web application firewall (WAF) service that enforces security policies at the edge of Google's network, in front of load balancers. It supports CEL-based custom rules that can inspect headers, query parameters, and request bodies to block specific malicious patterns like SQL injection, cross-site scripting, or known bot signatures. Because policies are evaluated before traffic reaches your GCE instances or GKE pods, attack traffic can be dropped with minimal latency impact, making it the correct choice for blocking a malicious traffic pattern.

Cloud Functions is the ideal fit because it is a managed Function-as-a-Service platform that executes code in response to a Pub/Sub event, scaling to zero when idle. It bills only for the actual invocation time, measured in 100ms increments, so a small Python snippet that runs once or twice undergoes no idle cost or container overhead. Its event-driven trigger model and built-in Pub/Sub subscription abstraction remove the need to run a web server or manage infrastructure, making it the most cost-effective and operationally simple choice for this task.

Taints on regular nodes with tolerations for stateful pods create an explicit scheduling contract: any pod without the matching toleration is blocked from the regular node pool, while stateful pods that declare the toleration are placed there exclusively. This ensures preemptible or stateless workloads cannot consume resources on those nodes, giving stateful services stable, non-preemptible infrastructure and predictable lifecycles.

BigQuery is a serverless data warehouse that uses columnar storage and a distributed query engine to run ANSI SQL analytics on petabytes of data without provisioning infrastructure. It separates storage and compute, allowing independent scaling, and introduces BigQuery ML to create and execute machine learning models directly on SQL queries. The service also integrates with Dataflow, Dataproc, and Looker for end-to-end analytics pipelines.

Compute Engine provides the virtual machines that form the worker and master nodes of a Cloud Dataproc cluster. When you run a Hadoop/Spark workload on Google Cloud, Cloud Dataproc orchestrates the deployment, but the actual CPU, memory, and local storage attached to each cluster node are Compute Engine instances. You can also run Hadoop/Spark directly on your own Compute Engine VMs without Dataproc, making Compute Engine the fundamental compute infrastructure for such workloads.

Cloud Functions is a serverless, event-driven compute service that executes code in response to specific triggers. It natively integrates with Google Cloud services such as Cloud Storage, Pub/Sub, and HTTP calls, automatically scaling the function instances as needed. You write and deploy a single-purpose function, and the platform handles the underlying infrastructure, so no server provisioning or cluster management is required. This makes it the direct answer for running code in response to events like a file upload or a message published to a topic.

Cloud CDN accelerates live video by caching video segments at Google's hundreds of edge PoPs, minimizing latency for global viewers even during stream spikes. Cloud Armor complements this by enforcing DDoS protection and security policies at the edge, blocking malicious traffic before it reaches the origin. This pairing delivers both performance and security, making it the correct solution for global live streaming.

Dataflow's unified streaming engine natively supports exactly-once processing via commit-and-finish plus its shuffle, and it provides event-time windowing and trigger strategies for late data. Its built-in BigQuery sink batches streaming records into load jobs, making this pipeline the recommended way to continuously ingest Pub/Sub events into BigQuery for clickstream analytics.

Cloud Run is the correct service because it directly executes any OCI-compliant container image on a fully managed, serverless infrastructure, eliminating the need to manage servers or clusters. It scales from zero to thousands of active instances in response to inbound HTTP requests, and charges only for the compute resources used while each request is being processed, plus a brief instance-startup window. This makes it ideal for a containerized web application: you retain portability and control over the runtime environment while benefiting from automatic TLS termination, revision traffic splitting, and the ability to scale down to literally zero when idle.

Preemptible VMs run on Google's surplus compute capacity and are available at up to 60–80% lower per-second cost than standard VMs. They can be reclaimed at any time and have a maximum runtime of 24 hours, making them ideal for idempotent, fault-tolerant batch processing. A nightly job can simply be restarted or resumed from a checkpoint if interrupted.

This plan matches lifecycle costs to actual access patterns. Product images are updated and viewed frequently in the first month, making Standard the low-cost choice; between day 30 and day 90, access drops to occasional reporting or past-order lookups, so Nearline reduces storage price while keeping retrieval fees reasonable; after 90 days, images become archival and rarely accessed, and Archive's roughly $0.0012/GB pricing is the cheapest option. Lifecycle rules automate both transitions, minimizing operational overhead.

A private pool in Cloud Build runs workers in a VPC network that you control, and these worker instances are provisioned without public IP addresses. Because they are internal-only, builds can pull source code from private repositories and push images to Artifact Registry without ever traversing the public internet or exposing the workers. This directly satisfies the requirement to remove public IPs while still allowing secure access to private resources.

Cloud Functions is a serverless Functions-as-a-Service platform with first-class support for Cloud Storage triggers (e.g., object finalize). When a video file is uploaded, a function is invoked automatically, scales from zero to handle the event, and charges only for execution time. No infrastructure provisioning or 24/7 VM is needed, making it the ideal lightweight, event-driven compute choice for this exact use case.

Cloud Load Balancing is the correct choice for distributing incoming HTTP traffic across backend instances or services. It provides global, anycast-based HTTP(S) load balancing with a single virtual IP, enabling traffic to be routed to the nearest healthy backend across regions. It also offers SSL/TLS offloading, autoscaling, health checks, and failover, making it the appropriate service for high-availability traffic distribution.

Cloud Run is a fully managed, serverless compute platform that executes stateless containers in response to HTTP requests. It automatically scales instances from zero up to handle traffic spikes and back down to zero when idle, so you pay only for the CPU and memory consumed during request processing—with no charge for idle-time zero-instance periods. Because Cloud Run abstracts away all infrastructure, you don't need to manage clusters, nodes, or virtual machines; you simply deploy a container image and let the service handle scaling, availability, and load balancing. This makes Cloud Run the ideal choice for a containerized web application with variable or intermittent traffic, providing minimal operational overhead and granular per-request billing.

Cloud Load Balancing, specifically the HTTP(S) Load Balancer, is a global Layer 7 solution that terminates SSL/TLS at Google's edge, distributes traffic across managed instance groups, and performs regular health checks to automatically route around failed backends. It supports content-based routing, autoscaling, and is a fully managed service.

Cloud Bigtable is a fully managed, wide-column NoSQL database built specifically for large-scale analytical and operational workloads, including time-series and IoT sensor data. It stores data as sparse rows keyed by a row key (typically device ID and timestamp), enabling single-digit millisecond read/write latency at massive scale. Bigtable scales horizontally by adding nodes to handle millions of queries per second without downtime, and its native integration with Cloud BigQuery, Dataflow, and Pub/Sub makes it the ideal choice for petabyte-scale sensor data ingestion and retrieval.

Vertex AI is Google Cloud's unified MLOps platform, designed to handle the entire ML lifecycle: data labeling, feature engineering, custom training with any framework (TensorFlow, PyTorch, etc.), hyperparameter tuning, model versioning, and serving through endpoints. It integrates services like Vertex AI Feature Store, Vertex AI TensorBoard, and Model Monitoring, enabling end-to-end management. For a data science team needing to train and deploy a custom model, Vertex AI provides the essential, scalable infrastructure.

Enabling session affinity (sticky sessions) on the load balancer ensures that all requests from a given client during a session are routed to the same backend instance, as long as that instance remains healthy. This preserves the in-memory session state because the application can store session data locally on the instance, and subsequent requests are consistently directed to that same machine. The load balancer typically uses a hash of the client's IP address or a generated cookie to determine the backend, while still balancing load across different sessions. This directly solves the problem of a stateful web application without needing to externalize or replicate session state.

Cloud Tasks decouples the image processing workload from the Cloud Function trigger, immediately returning a response to the client. Each upload is enqueued as a task that Cloud Run processes asynchronously, allowing you to configure timeouts up to 60 minutes or more. Cloud Run also scales to zero when idle, so you only pay for the processing time, and Cloud Tasks provides automatic retries for transient failures.

BigQuery is a serverless, fully managed data warehouse with columnar storage and a distributed execution engine, purpose-built for petabyte-scale SQL analytics. It separates compute from storage, enabling sub-second interactive queries over massive datasets without provisioning clusters. Its native support for standard SQL, partitioning, and clustering makes it the ideal service for analyzing petabytes of sales data.

Preemptible VMs offer up to an 80% discount compared to on-demand instances but may be terminated by Compute Engine at any time, with a maximum runtime of 24 hours. Since the nightly analytics job is batch-oriented and fault-tolerant, it can handle these interruptions via restart or checkpointing, making preemptible VMs the most cost-effective and appropriate choice. This aligns the compute pricing model with the workload's tolerance for interruption.

Bigtable is a fully managed, wide-column NoSQL database designed for massive scale, handling millions of writes per second with sub-10ms latency. It is ideal for high-throughput, time-series, or IoT data where each user profile is keyed by a row key and columns can be dynamically added. However, its strengths are in analytical and operational workloads with large volumes, not necessarily the best fit for a simple mobile app user profile, yet it works if data volume is extremely high and latency is critical.

Cloud Spanner is a fully managed, horizontally scalable relational database that combines standard SQL transactions with globally distributed replication. It uses TrueTime (a globally synchronized clock) and Paxos consensus to provide strong external consistency across all regions, meaning reads and writes are always linearizable even during failovers. With a 99.999% availability SLA and the ability to write from any region, it is the only option among these that supports globally consistent relational data at scale.

Pub/Sub provides a fully managed, asynchronous messaging service with durable message retention, at-least-once delivery, and replay support, enabling reliable, scalable stream ingestion with no operators. Cloud Dataflow, built on Apache Beam, processes streams in real time and supports event-time processing, watermarks, windowing, and exactly-once semantics, making it suitable for complex transformations, aggregations, and stateful analytics. As a fully managed service, it also autoscales and seamlessly integrates with GCP and open-source ecosystems, so this combination is the intended architecture for real-time stream processing on GCP.

Archive is the correct storage class because it is the lowest-cost option for long-term retention, designed specifically for data accessed less than once per year. It enforces a 365-day minimum storage duration, which aligns with archival retention needs where retrieval is rare and latency is acceptable. Its pricing model minimizes storage cost at the expense of higher retrieval fees, making it the most economical choice for compliance or forensic archives.

Cloud Armor provides built-in protections against volumetric DDoS attacks and a web application firewall (WAF) to block OWASP Top 10 threats like SQL injection and cross-site scripting. It uses preconfigured rules and adaptive protection to filter malicious traffic at the edge, only forwarding legitimate requests to backend services. This makes it the primary DDoS and application-layer security service for applications fronted by Cloud Load Balancing.

Cloud Armor is the correct choice because it provides web application firewall (WAF) capabilities, DDoS protection, and IP reputation filtering at the global edge, directly integrated with Cloud Load Balancing. It can inspect HTTP(S) traffic, block malicious requests based on Layer 7 attributes like headers and body, and enforce allow/deny lists based on IP addresses and geolocation. This gives the organization precise, policy-based control over incoming traffic before it reaches backend instances.

Preemptible VMs are instances that run on Google Cloud's surplus capacity at a fraction of the normal cost, often up to 80% cheaper. They can be terminated abruptly by Google Cloud, but a batch job that runs only once per day and can tolerate interruptions is an ideal candidate. By designing the job to restart or resume from checkpoints, the company can achieve substantial cost savings while maintaining reliability.

BigQuery is Google Cloud's serverless, fully managed data warehouse built on a columnar storage format and the Dremel massively parallel query engine, letting you run SQL over petabytes without provisioning. Its compute and storage are decoupled, so it scales elastically and you pay only for queries and stored data; native Looker integration and BI Engine in-memory acceleration directly support analytical workloads. That combination of serverless scalability, ANSI SQL, and built-in BI connectivity is what makes it the correct answer.

Cloud Firestore is a flexible, serverless NoSQL document database designed specifically for mobile and web clients, with real-time synchronization via listen callbacks and automatic offline data persistence. Its hierarchical data model organizes user profiles naturally as documents within collections, and security rules integrate directly with client SDKs. Unlike global SQL options, Firestore provides built-in multi-device sync and conflict handling, making it the ideal fit for a mobile app storing user profiles.

Preemptible VMs cost up to 80% less than standard VMs, making them ideal for fault-tolerant workloads that can survive instance termination, such as batch data processing, rendering, or stateless web serving. This cost benefit is the primary purpose of using them, not performance or availability, and it enables large-scale compute tasks under a constrained budget.

Firestore is a NoSQL document database with real-time synchronization built into its mobile SDKs, automatically pushing data changes to connected clients and persisting a local cache for offline operation. It resolves conflicts and syncs local writes when connectivity returns, enabling responsive, collaborative mobile experiences. Its flexible, schema-less document model maps naturally to app objects, making it the standard choice for real-time mobile apps.