Unit 4: Economic Benefits of Microservices - Practice Quiz

1 What is a key scalability advantage of microservices?

2 How can automatic scaling reduce cloud costs?

3 What is overprovisioning in cloud computing?

4 What does the pay-as-you-go cloud pricing model mean?

5 Which AWS pricing option provides unused compute capacity at a discounted price but may interrupt instances?

6 Which Azure pricing option can reduce costs when a customer commits to eligible resources for a fixed term?

7 What does Google Cloud offer for customers who commit to consistent resource usage for a specified term?

8 Which factor commonly affects the cost of a cloud virtual machine?

9 In serverless computing, customers are commonly charged based on what?

10 Which action can directly reduce compute costs for an unused virtual machine?

11 Which factor most directly determines the capacity charge for cloud storage?

12 Which storage class is generally most cost-effective for data that is rarely accessed?

13 Besides stored capacity, what may also generate object storage charges?

14 Which managed component can produce network-related cloud charges?

15 Which action can help reduce unnecessary network service costs?

16 Which type of cloud data transfer is commonly called egress?

17 Which data transfer direction is often free or less expensive with major cloud providers?

18 Which AWS tool helps users visualize and analyze their cloud spending?

19 Which Azure service is used to monitor, allocate, and optimize cloud spending?

20 What is the main purpose of a cloud budget alert?

21 An online retailer experiences heavy traffic for 4 hours each day and low traffic during the remaining 20 hours. Which microservices deployment strategy is likely to minimize compute cost while maintaining performance?

22 A document-conversion microservice receives requests only a few times per hour, and each conversion finishes within seconds. Which scaling approach offers the strongest cost advantage?

23 A company splits one application into 80 microservices, although most services always scale together. Which economic risk is most likely?

24 Which mapping correctly associates each cloud provider with a common discounted commitment model?

25 A batch-processing microservice can tolerate interruption and automatically retries unfinished jobs. Which pricing model is generally the most economical?

26 A microservice has stable baseline usage for the next three years but occasionally needs extra capacity. Which purchasing combination is usually most cost-effective?

27 A microservice runs on 6 instances costing $0.12 per instance-hour for 10 hours per day. What is its approximate 30-day compute cost?

28 A service uses 10 virtual machines at 20% average CPU utilization. Testing shows that 4 right-sized machines can meet the same performance target. If each machine costs $90 per month, what is the monthly compute saving?

29 A serverless microservice processes 4 million requests per month. Ignoring free tiers, its average compute charge is $0.00002 per request. What is the monthly compute charge?

30 A service stores 2,000 GB of data. It keeps 25% in standard storage at $0.023 per GB-month and 75% in archive storage at $0.004 per GB-month. What is the monthly storage cost?

31 A team moves old logs to a low-cost archive tier but frequently queries them for troubleshooting. Which cost is most likely to reduce the expected savings?

32 A microservice retains all application logs indefinitely in high-performance storage. Which change would most directly lower storage cost without immediately deleting recent logs?

33 Several microservices exchange large messages through a managed network gateway even though they run in the same cloud region. Which design change should be evaluated first to reduce network cost?

34 A public API repeatedly serves the same large images to users around the world. Which architecture is most likely to reduce origin network traffic and improve latency?

35 A cloud provider charges $0.09 per GB for the first 1,000 GB of monthly outbound traffic and $0.07 per GB for the next tier. What is the charge for 5,000 GB of outbound traffic?

36 A frontend in one cloud region frequently calls a backend in another region, transferring several terabytes each month. Which change most directly targets the resulting data transfer cost?

37 A microservice sends 10 TB of uncompressed JSON responses to external clients each month. Which change is most likely to reduce outbound data transfer charges?

38 A team wants an alert when projected monthly cloud spending is likely to exceed $5,000. Which capability should it configure?

39 A cloud bill rises unexpectedly, but total user traffic remains stable. Which tool capability is most useful for quickly identifying the unusual spending pattern?

40 A company cannot determine which microservice is responsible for shared cloud costs. Which practice would provide the most useful cost attribution?

41 A monolith contains services A and B. One monolith instance costs $0.18/hour and can handle 100 A requests/s and 20 B requests/s. At peak, A receives 600 requests/s and B receives 60 requests/s. If separated, an A instance costs $0.12/hour with capacity 100 requests/s, while a B instance costs $0.08/hour with capacity 20 requests/s. Ignoring shared overhead, what is the hourly cost effect of independent scaling?

42 A service needs 4 instances normally and 20 instances during a 10-minute burst each hour. Sixteen additional instances take 5 minutes to start and remain allocated for a 10-minute cooldown after the burst ends. Assuming per-minute billing begins when an instance becomes available, how does autoscaling compare with keeping 20 instances active for the entire hour?

43 At traffic multiplier , a monolith requires instances, while independently scalable microservices require instances. Each instance costs $50/month, but the microservices platform adds a fixed $2,800/month for gateways, observability, and control-plane services. If must be a positive integer, what is the smallest value at which microservices are strictly cheaper?

44 A company has a stable aggregate compute spend but frequently moves workloads among EC2 instance families and also shifts some services between EC2, AWS Fargate, and AWS Lambda. Which AWS pricing model most directly preserves commitment discounts across those changes?

45 An organization runs steady Windows Server and SQL Server workloads on Azure and owns eligible licenses with Software Assurance. Which strategy generally minimizes the stable baseline cost without relying on interruptible capacity?

46 A Google Cloud workload continuously uses 60 vCPUs and bursts to 100 vCPUs for 12 full days each month. The VMs are eligible for sustained-use discounts. Which purchasing strategy best avoids overcommitment while targeting both predictable and variable usage?

47 An on-demand instance costs $0.40/hour, while an equivalent Spot instance costs $0.12/hour. If interruptions make a fraction of all paid Spot execution unusable, checkpoint overhead is otherwise negligible. For what range of is Spot still strictly cheaper per useful compute-hour?

48 A serverless microservice handles 80 million requests per month. Each request uses 1.5 GB of memory for an average of 0.24 seconds. Pricing is $0.000016667 per GB-second plus $0.20 per million requests, with free tiers ignored. What is the approximate monthly compute and request cost?

49 A Kubernetes node provides 8 vCPUs and 32 GB of memory for $0.40/hour. The workload has twenty pods requesting 1 vCPU and 2 GB each, plus eight pods requesting 0.5 vCPU and 8 GB each. An anti-affinity rule permits at most one of the 8-GB pods per node. Ignoring system reservations, what is the minimum node cost per hour?

50 A system stores 100 TB using decimal units. Standard storage costs $0.023/GB-month. Archive storage costs $0.004/GB-month plus $0.03 per GB retrieved. Ignoring request, minimum-duration, and transition charges, below what monthly retrieval fraction $f$ is archive storage strictly cheaper?

51 A microservice stores 60 TB of logical data. Triple replication uses three times the logical capacity. An erasure-coded design uses 1.5 times the logical capacity but adds $450/month in compute overhead. Storage costs $0.02/GB-month using decimal units. What saving does erasure coding provide?

52 An object store contains 8 TB of current data. Each day, 2% of the dataset is replaced, and every replaced portion remains as a noncurrent version for exactly 30 days. Assuming a constant dataset size, no compounding, decimal units, and storage pricing of $0.021/GB-month, what is the steady-state monthly storage cost?

53 Two chatty microservices run in different availability zones. Monthly metering records 12,000 GB sent from service A to B and 12,000 GB returned from B to A. Cross-zone traffic costs $0.01/GB in each direction. What incremental monthly charge could co-location eliminate, ignoring resilience effects?

54 Private workloads transfer 40 TB per month to object storage through a NAT gateway. The gateway charges $0.045/GB processed plus $0.045/hour for 730 hours. A private gateway endpoint has no hourly or per-GB processing charge for this traffic. Using decimal units and ignoring unrelated charges, what monthly amount is avoided?

55 Direct origin delivery costs $0.09/GB. A CDN costs $0.07/GB delivered to users, and each cache miss adds $0.04/GB for origin-to-CDN transfer. If $h$ is the cache-hit ratio and request charges are ignored, when is the CDN strictly cheaper?

56 A cloud provider charges monthly internet egress at $0.09/GB for the first 10 TB, $0.085/GB for the next 40 TB, and $0.07/GB thereafter. A system sends 60 TB in a month using decimal units. What is its total egress charge?

57 A multi-region service ingests 5 TB per month from the internet into region A, synchronously replicates all of it once to region B, and serves 3 TB per month from region B to internet clients. Internet ingress is free, inter-region transfer costs $0.02/GB, and internet egress costs $0.08/GB. Using decimal units, what is the monthly transfer cost?

58 An AWS account prepays $3,650 for a one-year commitment on the first day of a 365-day term. A FinOps team wants a 30-day report that attributes the commitment evenly across its useful period rather than showing the cash purchase entirely on day one. Which view and amount are appropriate?

59 A shared observability cluster has no reliable per-team infrastructure tags, but its telemetry records log-ingestion bytes and metric-series counts by tenant. Which cost-allocation approach gives the most defensible showback?

60 A rightsizing tool evaluates an 8-vCPU instance costing $0.40/hour. Its average, p95, and p99 CPU utilization are 20%, 55%, and 92%. The SLO requires projected p95 utilization not to exceed 80%, while brief p99 throttling is acceptable. Candidate instances have 6 vCPUs at $0.32/hour and 4 vCPUs at $0.22/hour. Assuming utilization scales inversely with vCPU count, what is the best recommendation for a 730-hour month?