Download Speaker Cloud Native Sustainability Week Practice Exam and more Exams Technology in PDF only on Docsity!
Practice Exam
Question 1. Which principle defines the practice of designing software to minimize its environmental impact throughout its lifecycle? A) Agile Development B) Green Software Engineering C) DevSecOps D) Continuous Delivery Answer: B Explanation: Green Software Engineering focuses on reducing carbon emissions, energy use, and waste from design to disposal. Question 2. What does the term “embodied carbon” refer to in the context of IT infrastructure? A) Carbon emitted during runtime of workloads B) Carbon released when disposing of hardware C) Carbon associated with manufacturing and transporting hardware D) Carbon generated by cooling systems only Answer: C Explanation: Embodied carbon accounts for emissions from raw material extraction, manufacturing, and logistics of hardware components. Question 3. Which metric measures the amount of carbon emitted per unit of energy consumed by a data center? A) Power Usage Effectiveness (PUE) B) Carbon Usage Effectiveness (CUE) C) Energy Use Intensity (EUI) D) Data Center Efficiency Ratio (DCER) Answer: B
Practice Exam
Explanation: CUE = total carbon emissions / total energy used, indicating carbon efficiency of a data center. Question 4. In cloud sustainability, Scope 2 emissions are primarily associated with which of the following? A) Direct fuel combustion on‑premises B) Purchased electricity from the grid C) Upstream supply‑chain activities D) End‑user device usage Answer: B Explanation: Scope 2 covers indirect emissions from electricity, heat, or steam purchased from external providers. Question 5. Which of the following best describes the shared responsibility model for sustainability? A) Cloud providers handle all emissions; consumers have no role B) Consumers must offset all emissions themselves C) Providers supply efficient infrastructure while consumers optimize workloads D) Only developers need to consider carbon footprints Answer: C Explanation: Sustainability is a joint effort: providers deliver low‑carbon infrastructure, and consumers design efficient applications. Question 6. The Green Software Foundation’s “Energy” principle emphasizes which practice? A) Using only renewable energy sources for development laptops B) Measuring and reducing the energy consumption of software at runtime
Practice Exam
B) Lower overall server utilization C) Access to renewable‑energy‑powered data centers and better cooling efficiency D) Mandatory use of proprietary APIs Answer: C Explanation: Public clouds achieve high utilization, optimized cooling, and often source power from renewables. Question 10. Power Usage Effectiveness (PUE) is calculated as: A) Total Facility Energy / IT Equipment Energy B) IT Equipment Energy / Total Facility Energy C) Carbon Emissions / Energy Consumed D) Energy Consumed / Compute Hours Answer: A Explanation: PUE = (Total Facility Energy) ÷ (IT Equipment Energy); lower values indicate higher efficiency. Question 11. Which factor does NOT directly influence a workload’s total carbon footprint in a cloud environment? A) Power consumed by the underlying hardware B) Carbon intensity of the regional electricity grid C) Number of source code lines in the application D) Embodied carbon of the servers running the workload Answer: C Explanation: Source‑code size does not directly affect carbon emissions; hardware usage and energy sources do.
Practice Exam
Question 12. Kepler, a Kubernetes‑based Efficient Power Level Exporter, primarily provides which type of data? A) Application latency metrics B) Per‑pod power consumption estimates C) Network throughput statistics D) Security vulnerability scans Answer: B Explanation: Kepler collects and exports power usage metrics at the pod/container level for carbon accounting. Question 13. Which open‑source stack is commonly used to visualize resource utilization and can be extended to display carbon metrics? A) ELK Stack (Elasticsearch, Logstash, Kibana) B) Prometheus + Grafana C) Splunk Enterprise D) Nagios Core Answer: B Explanation: Prometheus scrapes metrics while Grafana visualizes them; both can integrate carbon data from tools like Kepler. Question 14. The AWS Customer Carbon Footprint Tool provides: A) Real‑time CPU throttling alerts B) Detailed per‑service carbon emission reports for an AWS account C) Automatic container image optimization D) Network packet loss diagnostics Answer: B
Practice Exam
D) LowPriority Answer: C Explanation: Guaranteed pods have both CPU and memory requests equal to limits, giving them top scheduling priority. Question 18. Right‑sizing a workload involves: A) Increasing its resource limits arbitrarily B) Reducing its code complexity C) Analyzing utilization data and adjusting requests/limits to match actual usage D) Switching from containers to VMs Answer: C Explanation: Right‑sizing uses observed metrics to set appropriate resource specifications, minimizing waste. Question 19. Horizontal Pod Autoscaler (HPA) can scale based on which type of metric? A) Only CPU utilization B) Only memory usage C) Custom or external metrics such as request latency or queue length D) Number of Git commits Answer: C Explanation: HPA supports built‑in metrics (CPU, memory) and custom/external metrics via the Metrics API. Question 20. Vertical Pod Autoscaler (VPA) differs from HPA because it: A) Adds more pod replicas instead of changing resources per pod B) Adjusts the CPU/memory requests of existing pods rather than replica count
Practice Exam
C) Only works on Windows nodes D) Requires manual intervention for each scaling event Answer: B Explanation: VPA recommends or applies changes to a pod’s resource requests/limits, optimizing per‑pod sizing. Question 21. The Kubernetes Cluster Autoscaler primarily helps to: A. Increase the number of namespaces automatically B. Add or remove nodes based on pending pod resource requests C. Scale individual containers without affecting pods D. Manage persistent volume claims size Answer: B Explanation: Cluster Autoscaler monitors unscheduled pods and adjusts node count to meet demand. Question 22. KEDA (Kubernetes Event‑Driven Autoscaling) enables scaling to zero for workloads that are: A. CPU‑bound continuously B. Event‑driven and idle most of the time C. Stateless but always active D. Running on dedicated hardware Answer: B Explanation: KEDA watches external event sources and can scale workloads down to zero when no events are present. Question 23. Which practice helps to eliminate orphaned resources in a Kubernetes cluster?
Practice Exam
Question 26. Multi‑stage Docker builds contribute to sustainability by: A. Increasing the number of layers in the final image B. Reducing the final image size, lowering storage and transfer energy C. Automatically scaling containers based on load D. Embedding secret keys in the image Answer: B Explanation: Multi‑stage builds allow removal of build‑time dependencies, creating leaner images that consume less storage and bandwidth. Question 27. Carbon‑aware scheduling aims to: A. Run all workloads at the same time regardless of grid conditions B. Prefer nodes in regions with lower real‑time carbon intensity for non‑urgent jobs C. Disable autoscaling in high‑carbon regions D. Increase CPU frequency during peak renewable generation Answer: B Explanation: By aligning compute with low‑carbon grid periods, overall emissions are reduced. Question 28. Which source can provide real‑time regional carbon intensity data for scheduling decisions? A. Local DNS resolver logs B. Carbon intensity APIs such as electricityMap.org C. Container image tags D. Git commit timestamps Answer: B
Practice Exam
Explanation: Public APIs publish grid carbon intensity, enabling applications to make carbon‑aware choices. Question 29. An energy‑efficient algorithm typically aims to: A. Maximize the number of loops regardless of work done B. Reduce computational complexity and avoid unnecessary I/O C. Use the highest possible memory allocation D. Increase network chatter to keep connections alive Answer: B Explanation: Lower algorithmic complexity and minimized I/O lower CPU cycles and energy consumption. Question 30. Which design pattern can help reduce database load and thus lower energy use? A. Singleton per request B. Aggressive polling C. Caching frequently accessed data D. Storing all data in in‑memory structures only Answer: C Explanation: Caching reduces repetitive database queries, cutting compute and I/O energy. Question 31. When selecting a programming language for a new microservice, which factor influences its carbon footprint the most? A. Popularity on GitHub B. Runtime efficiency and memory usage of the language’s interpreter/VM C. Number of available third‑party libraries D. Syntax similarity to JavaScript
Practice Exam
D. SMTP
Answer: B Explanation: gRPC uses binary payloads and multiplexed streams, reducing overhead and bandwidth. Question 35. Lifecycle management of stored data (archiving, deletion) helps reduce carbon emissions by: A. Increasing the number of active disks B. Keeping all data hot for instant access C. Lowering storage utilization and associated power draw D. Replicating data across all zones Answer: C Explanation: Removing or archiving unused data frees storage resources, decreasing power consumption. Question 36. The CNCF TAG ENV (Technical Advisory Group – Environmental Sustainability) focuses on: A. Defining networking protocols for IoT B. Developing sustainability standards and best practices for cloud‑native ecosystems C. Certifying Kubernetes distributions for security only D. Managing CNCF financial assets Answer: B Explanation: TAG ENV works on sustainability guidelines, metrics, and community collaboration. Question 37. Contributing code to the Kepler project can help an organization: A. Reduce licensing costs for proprietary software
Practice Exam
B. Directly influence the accuracy of energy measurement tools used internally C. Avoid using any monitoring solutions altogether D. Increase the number of containers per node beyond hardware limits Answer: B Explanation: By improving Kepler, organizations gain more precise power data, aiding carbon reduction efforts. Question 38. Incorporating sustainability goals into the SDLC typically involves: A. Ignoring performance testing to save time B. Adding carbon‑impact assessments to design reviews and CI pipelines C. Removing all security checks to speed up deployment D. Using only on‑premises hardware for development Answer: B Explanation: Embedding carbon metrics into design, testing, and CI ensures sustainability is evaluated continuously. Question 39. “Green FinOps” treats carbon as a first‑class metric alongside cost and performance. Which practice exemplifies this approach? A. Ignoring carbon data during budgeting B. Allocating budget based on both monetary cost and estimated emissions per workload C. Only tracking CPU utilization for billing D. Using fixed‑price contracts without sustainability clauses Answer: B Explanation: Green FinOps integrates carbon accounting into financial decision‑making, balancing cost, performance, and emissions.
Practice Exam
Explanation: Multi‑cloud enables workload placement in regions powered by cleaner energy sources. Question 43. In the context of Kubernetes, “idle” pods contribute to carbon emissions because: A. They automatically shut down when not receiving traffic B. Their allocated resources remain powered even without work C. They are always terminated after each request D. They run on serverless platforms only Answer: B Explanation: Pods consume CPU/memory resources continuously unless scaled down or terminated. Question 44. Which of the following is NOT a recommended practice for container image optimization? A. Using minimal base images like Alpine B. Removing build‑time dependencies via multi‑stage builds C. Including large debugging tools in the final image D. Compressing layers during build Answer: C Explanation: Shipping unnecessary tools inflates image size, increasing storage and transfer energy. Question 45. The “Carbon Usage Effectiveness (CUE)” metric is expressed as: A. Kilowatt‑hours per unit of compute B. Kilograms of CO₂e per kilowatt‑hour of total facility energy C. Dollars per CPU core
Practice Exam
D. Megabytes transferred per second Answer: B Explanation: CUE = total CO₂e emissions ÷ total energy consumption (kWh), indicating carbon efficiency. Question 46. Which of the following actions directly reduces Scope 1 emissions for a cloud‑native application? A. Purchasing renewable energy certificates (RECs) for the data center B. Optimizing code to use fewer CPU cycles C. Deploying to a provider that uses renewable‑powered data centers D. Offloading work to edge devices powered by diesel generators Answer: C Explanation: Scope 1 covers direct emissions from owned facilities; using a provider with renewable energy reduces those emissions. Question 47. What does “burstable” QoS mean for a pod’s scheduling behavior? A. It receives guaranteed resources at all times B. It can use additional CPU when node resources are idle, but may be evicted under pressure C. It never gets scheduled on a node with other pods D. It is always placed in a dedicated node pool Answer: B Explanation: Burstable pods have lower priority; they can exceed requests when capacity allows but may be throttled or evicted. Question 48. Which metric would you monitor to detect over‑provisioned memory in a Kubernetes cluster?
Practice Exam
Question 51. Which of the following best describes “embodied carbon” in the context of a container image? A. CO₂ emitted during image pull over the network B. CO₂ associated with the storage media and data center infrastructure that host the image C. CO₂ produced by the CPU while the container runs D. CO₂ released when the image is deleted Answer: B Explanation: Embodied carbon includes emissions from manufacturing and operating the storage hardware that holds the image. Question 52. Which practice helps reduce network‑related carbon emissions for a microservice architecture? A. Using verbose JSON payloads for every request B. Implementing gRPC with protobuf to minimize data size and connection overhead C. Deploying each service in a separate geographic region regardless of data source D. Enabling debug logging on all production traffic Answer: B Explanation: Efficient binary protocols reduce bandwidth and processing, lowering energy consumption. Question 53. Which of the following is a direct outcome of right‑sizing pods using VPA recommendations? A. Increased number of pod restarts due to memory pressure B. Lowered CPU throttling and better node utilization C. Automatic migration of workloads to a different cloud provider D. Disabling of Horizontal Pod Autoscaling
Practice Exam
Answer: B Explanation: VPA aligns pod resources with actual usage, reducing throttling and improving overall utilization. Question 54. In a multi‑cloud environment, which tool can aggregate carbon emissions across providers for unified reporting? A. Kubernetes Dashboard B. Cloud Carbon Footprint (CCF) C. Istio Service Mesh D. Helm Chart Repository Answer: B Explanation: CCF collects usage data from various clouds and calculates combined emissions. Question 55. Which of the following actions would most likely increase the PUE of a data center? A. Improving cooling efficiency through liquid cooling B. Consolidating workloads onto fewer servers C. Adding unused idle servers to the facility D. Using renewable energy for power supply Answer: C Explanation: Adding idle hardware raises total facility energy without increasing IT energy, worsening PUE. Question 56. Which component of the Green Software Foundation’s principles focuses on the design of hardware to be more sustainable? A. Carbon
