PCD Timed Hard Questions — 25 Difficult Questions

Question 1

Your application uses Cloud Spanner with strong reads, but you are experiencing high latency for read requests. You don't need absolute consistency for every read; stale data up to 5 seconds is acceptable. How can you reduce read latency?

15 minute time limit — choose an answer to begin.

Accepted Answer

Use bounded staleness reads with a max staleness of 5 seconds. Using bounded staleness allows reads to be served from replicas that may be slightly stale (up to 5 seconds), reducing latency. Read timestamps and DML are not relevant, and global indexes would not help.

Answer

Use the read timestamp to read the latest data

Answer

Create a global secondary index on frequently read columns

Answer

Use DML instead of mutations for writes

Question 2

A company uses Cloud Run for a serverless application that processes user uploads. Users report that sometimes the first request after a period of inactivity takes very long (cold start). The application is stateless. They want to minimize cold start latency while keeping costs low. The application is deployed with default settings: min instances = 0, max instances = 100, CPU always off, and a container image of 1GB. What should they do to reduce cold start latency?

Accepted Answer

Set min instances to 1 to keep a warm instance.. Setting min instances to 1 ensures that at least one instance is always warm and ready to serve requests, eliminating the cold start for the first request after a period of inactivity. Since the application is stateless and the default min instances is 0, Cloud Run scales down to zero, causing a cold start on the next request. By keeping one instance warm, you minimize latency without significantly increasing cost, as you only pay for the single idle instance.

Answer

Increase container memory from the default to reduce startup time.

Answer

Use a larger container image to include more dependencies.

Answer

Enable CPU always on allocation.

Question 3

A team is deploying a microservices application on Cloud Run and needs to implement canary deployments with traffic splitting. They are using Cloud Deploy. What is the correct configuration to gradually shift traffic from the old revision to the new revision?

Accepted Answer

Use a Cloud Deploy delivery pipeline with a canary strategy that specifies percentages like [5, 10, 50, 100] and includes a verification step.. Cloud Deploy natively supports canary deployments with traffic splitting for Cloud Run. By defining a canary strategy with incremental percentages (e.g., [5, 10, 50, 100]) and including a verification step, the pipeline automatically shifts traffic in stages, pausing for verification at each phase to ensure the new revision is healthy before progressing. This approach integrates directly with Cloud Deploy's delivery pipeline, eliminating the need for manual scripts or external API calls.

Answer

Use Cloud Build to deploy with a script that gradually increases traffic using the Cloud Run API.

Answer

Use a Cloud Deploy pipeline with a blue-green strategy that swaps all traffic at once.

Answer

Use Cloud Run's built-in traffic splitting with `gcloud run deploy --traffic` and manage manually.

Question 4

A financial trading application on Compute Engine requires an RPO of 5 seconds and RTO of 1 minute for zone failures. Which architecture should they use?

Accepted Answer

Regional persistent disk attached to a single instance. Regional persistent disks provide synchronous replication of data between two zones within a region, ensuring an RPO of effectively zero (typically under 5 seconds) and enabling rapid failover to a secondary zone. By attaching the regional disk to a single Compute Engine instance, the application can quickly resume operations in the other zone upon failure, meeting the 1-minute RTO without data loss or complex replication overhead.

Answer

Persistent disk with periodic snapshots to a different zone

Answer

Managed instance group with autoscaling and health checks

Answer

Two instances in different zones with data replicated via rsync

Question 5

A company is using Cloud Deploy to manage canary deployments to GKE. They want to automatically promote a release to the 'production' target if the canary deployment in the 'staging' target passes a set of automated smoke tests. What is the required configuration?

Accepted Answer

Define a deployment verifier in the pipeline that runs smoke tests and promotes on success.. Cloud Deploy supports deployment verifiers, which are custom Cloud Build jobs that run as part of a rollout. By defining a verifier in the pipeline that executes automated smoke tests, the canary deployment in the staging target can be automatically promoted to production only if the verifier succeeds. This integrates testing directly into the delivery pipeline without manual intervention.

Answer

Create a Cloud Build trigger to redeploy on test success.

Answer

Configure a manual approval gate between staging and production in the delivery pipeline.

Answer

Set the 'automaticPromotion' flag to true on the staging target.

Question 6

A company is migrating a monolithic application to microservices on Google Cloud. They have strict requirements for service-to-service communication: requests must be authenticated, authorized, and encrypted in transit. They also need to enforce fine-grained access control based on the requesting service identity. Which Google Cloud service should they use to achieve these goals?

Accepted Answer

Cloud Service Mesh (Anthos Service Mesh). Cloud Service Mesh (Anthos Service Mesh) provides mutual TLS, fine-grained authorization policies, and service identity. Option A is for API management, option B is for DDoS protection, and option D is for serverless ingress control.

Answer

Cloud Run with ingress control

Answer

Cloud Armor with IAM

Answer

Cloud Endpoints with API keys

Question 7

A company running a high-traffic e-commerce platform on Google Cloud experiences occasional data loss in their Cloud SQL database during failover events. The database is configured with a failover replica in a different zone. What is the most likely cause of the data loss?

Accepted Answer

The database is using asynchronous replication to the failover replica.. Cloud SQL uses synchronous replication for failover replicas by default, ensuring that transactions are committed on both the primary and the replica before acknowledging the write. If asynchronous replication is configured, the replica may lag behind the primary, and during a failover, any transactions not yet replicated are lost. This is the most likely cause of data loss during failover events.

Answer

Automated backups are not enabled.

Answer

The failover replica is configured as a read replica instead of a failover replica.

Answer

The database is not using regional persistent disks.

Question 8

A company is using Cloud Monitoring to track custom metrics published from an on-premises application using the Monitoring API. The metrics are published every 30 seconds. The team wants to create an alert that fires if the metric goes below a threshold for more than 1 minute. Which alert condition type should they use?

Accepted Answer

Metric type: custom.googleapis.com, condition: metric threshold (below value), duration: 60s. The requirement is to fire an alert when the metric value goes below a threshold for more than 1 minute. The 'metric threshold (below value)' condition type directly monitors a metric against a lower bound and evaluates the condition over the specified duration (60s) before triggering. This aligns with the need to detect sustained low values, not just a single data point.

Answer

Metric type: custom.googleapis.com, condition: metric lower bound, duration: 60s

Answer

Metric type: custom.googleapis.com, condition: metric change, duration: 60s

Answer

Metric type: custom.googleapis.com, condition: metric absence, duration: 60s

Question 9

A company runs a microservices application on Google Kubernetes Engine. They use Cloud SQL for persistent data. Recently, during a traffic spike, the application experienced increased latency and some requests failed with timeout errors. The team observed that the Cloud SQL CPU utilization spiked to 100%, and the GKE pods had high memory usage. They are using a standard Cloud SQL tier (db-n1-standard-2). Which course of action would best improve the application's performance and reliability?

Accepted Answer

Implement caching with Memorystore for frequently accessed data.. The primary bottleneck is the Cloud SQL CPU spiking to 100% under heavy read traffic. Implementing Memorystore (Redis) caching offloads repeated read queries from the database, reducing CPU load and query latency. This directly addresses the root cause—database CPU exhaustion—without requiring a larger database instance or adding replicas that would still be limited by the same CPU.

Answer

Upgrade Cloud SQL to a higher tier with more CPU.

Answer

Increase the number of replicas in GKE to reduce load per pod.

Answer

Add read replicas to Cloud SQL.

Question 10

An engineer is designing a Cloud Spanner schema for a social media application. The Users table and the Posts table have a parent-child relationship. To optimize read performance for fetching all posts of a user, which schema design approach should be used?

Accepted Answer

Use interleaved tables with Users as parent and Posts as child. Interleaved tables store child rows (Posts) with their parent row (User) in the same split, enabling locality and faster joins. Global or local indexes are separate structures.

Answer

Denormalize by storing post data as a repeated field in Users table

Answer

Use a local secondary index on user_id in the Posts table

Answer

Create a global secondary index on user_id in the Posts table

Question 11

You are designing a serverless application using Cloud Functions that processes events from Cloud Storage and Cloud Pub/Sub. The function must be idempotent and handle duplicate events. Which three best practices should you implement? (Choose THREE.)

Accepted Answer

Generate a unique idempotency key for each event and store processed keys in a database.. Generating a unique idempotency key for each event and storing processed keys in a database (such as Cloud Firestore) ensures that if the same event is delivered multiple times (e.g., due to at-least-once delivery semantics in Cloud Pub/Sub or Cloud Storage notifications), the function can check the key before processing and skip duplicates. This pattern is essential for idempotent serverless functions, as Cloud Functions may be retried on failure or receive duplicate events from the source.

Answer

Invoke the function synchronously to avoid duplicates.

Answer

Set the function timeout to maximum (540 seconds) to ensure processing completes.

Question 12

An application running on GKE uses a custom metric to track order processing time. The metric is exported via Prometheus and ingested by Cloud Monitoring using the Managed Service for Prometheus. The team wants to create an alert when the 95th percentile latency exceeds 2 seconds over a 5-minute window. Which PromQL query should be used?

Accepted Answer

histogram_quantile(0.95, rate(order_processing_duration_seconds_bucket[5m])). `histogram_quantile(0.95, rate(order_processing_duration_seconds_bucket[5m]))` computes the 95th percentile latency over a 5-minute window using Prometheus histogram buckets. The `rate()` function calculates the per-second increase of each bucket, which is required for accurate quantile estimation from cumulative histograms, and the result directly gives the latency threshold below which 95% of requests fall.

Answer

avg(rate(order_processing_duration_seconds_sum[5m])) / avg(rate(order_processing_duration_seconds_count[5m]))

Answer

histogram_quantile(0.95, sum(rate(order_processing_duration_seconds_bucket[5m])))

Answer

histogram_quantile(0.95, order_processing_duration_seconds_bucket)

Question 13

An organization has Cloud Bigtable for real-time ad serving and BigQuery for analytics. They want to synchronize data changes from Bigtable to BigQuery every hour with a Dataflow pipeline. However, they notice that some updates are missed. What is the most likely cause?

Accepted Answer

Bigtable does not support change data capture natively. Bigtable does not have built-in change data capture (CDC) like some databases. Dataflow can read Bigtable snapshots but not incremental changes unless using Dataflow's streaming insert, which is not native. The common approach is to use Bigtable replication or export, but missed updates often occur because Bigtable does not log changes. A custom solution using Cloud Bigtable change streams (beta) or using a separate logging table is needed.

Answer

The Dataflow pipeline reads from Bigtable using a snapshot read

Answer

The Bigtable table has no primary key

Answer

The Dataflow pipeline is using the wrong sink

Question 14

A developer is designing a chat application using Cloud Firestore. They need to ensure that updates to messages are propagated to all clients in real-time. Which feature should they use?

Accepted Answer

Real-time listeners. Real-time listeners (onSnapshot) in Cloud Firestore allow clients to subscribe to document or query changes, receiving updates immediately when data is modified. This ensures all connected clients see message updates in real-time without polling, which is essential for a chat application.

Answer

Firestore indexes

Answer

Security rules

Answer

Offline persistence

Question 15

Refer to the exhibit. A developer runs the above command to deploy a Cloud Function triggered by Pub/Sub. The function fails to execute when a message is published. The logs show: "Function execution took 60001 ms, finished with status: 'timeout'". What should the developer do?

Accepted Answer

Check the function code for long-running operations. The timeout error indicates the Cloud Function is exceeding its maximum execution duration. The default timeout for Cloud Functions is 60 seconds, and the logs confirm the function ran for 60001 ms before being forcibly terminated. The most likely cause is that the function code contains long-running operations (e.g., synchronous HTTP calls, database queries, or heavy computation) that do not complete within the allotted time. Therefore, the developer should inspect and optimize the function code to reduce execution time, such as by using asynchronous processing or breaking the work into smaller chunks.

Answer

Change the trigger to HTTP

Answer

Reduce the number of function instances

Answer

Increase the function timeout to 9 minutes

Question 16

You have a Cloud Bigtable instance with a single table that stores user events. You need to ensure that data older than 30 days is automatically deleted. What should you configure?

Accepted Answer

Set a garbage collection policy on the default column family to delete cells older than 30 days. Bigtable garbage collection policies operate on column families and can delete cells older than a specified time or keep a max number of versions.

Answer

Set a time-to-live (TTL) on the table

Answer

Use a cron job to manually delete rows with timestamps older than 30 days

Answer

Enable compaction and set a TTL on the table

Question 17

A company is migrating an on-premises PostgreSQL database to Cloud SQL for PostgreSQL using Database Migration Service (DMS). During the continuous migration phase, the source database has high write throughput, and the DMS job is falling behind. The replication lag is increasing over time. What should the engineer do to reduce lag?

Accepted Answer

Increase the wal_keep_segments parameter on the source and enable parallel apply on the destination Cloud SQL instance. Increasing wal_keep_segments on the source ensures that more WAL segments are retained, preventing DMS from failing to fetch required logs due to premature cleanup. Enabling parallel apply on the Cloud SQL destination allows multiple transactions to be applied concurrently, reducing the time needed to catch up with the source's high write throughput.

Answer

Change the source database to use synchronous replication with DMS

Answer

Reduce the maximum WAL size (max_wal_size) on the source to generate smaller logs

Answer

Increase the maintenance_work_mem parameter on the source database

Question 18

A company uses Cloud SQL for MySQL and wants to achieve high availability with automatic failover across zones while minimizing data loss. Which configuration should they use?

Accepted Answer

Configure a regional Cloud SQL instance with automatic failover. A regional Cloud SQL instance with automatic failover provides synchronous replication of data between two zones within the same region, ensuring zero data loss (RPO=0) and automatic failover with minimal downtime (RTO typically under 60 seconds). This meets the requirement for high availability with automatic failover across zones while minimizing data loss.

Answer

Enable read replicas in different zones

Answer

Use external read replicas with a failover script

Answer

Use Cloud SQL Enterprise Plus edition

Answer

Enable point-in-time recovery

Question 19

A multi-region application uses Cloud Spanner. The team needs to ensure that a write is immediately visible to all subsequent reads, even those performed in different regions. Which consistency mode should they use?

Accepted Answer

Strong consistency. Strong consistency (D) ensures that once a write is acknowledged, any subsequent read, regardless of region, will reflect that write. Cloud Spanner uses the TrueTime API and Paxos-based replication to provide external consistency (a form of strong consistency) across regions, making it the correct choice for immediate global visibility.

Answer

Eventual consistency

Answer

Global consistency

Answer

Bounded staleness

Question 20

You want to configure Cloud SQL for MySQL with high availability (HA). The application requires that failover be automatic and that the failover replica is in a different zone within the same region. Which configuration should you use?

Accepted Answer

Enable the high availability option when creating the Cloud SQL instance. Enabling the high availability (HA) option when creating a Cloud SQL for MySQL instance automatically provisions a standby replica in a different zone within the same region. This configuration provides automatic failover without manual intervention, meeting the requirement for HA with zone-level redundancy.

Answer

Use Cloud SQL with a cross-region replica and configure failover with gcloud

Answer

Set up a Cloud SQL instance with multiple read replicas and use failover via the Cloud SQL Proxy

Answer

Create a read replica in a different zone and configure automatic failover using a load balancer