Deep Dive 27.07.2025

Regarding memory cgroups, hard limits, and customized solutions for overriding these limits, which of the following statements are correct?

Which of the following statements are true regarding span attributes, handling the trace ID, and annotations in Dapper-style distributed tracing?

Consider five different IoT network technologies: Bluetooth, ZigBee, Wi-Fi, LoRa, and LTE. Describe a specific application scenario for each technology, justifying your choice based on the technology's range, maximum transmission unit (MTU), and data transfer rate. Explain why the chosen technology is most suitable for that particular application, considering the trade-offs between these three parameters.

Analyze the relationship between a WebAssembly (WASM) module and a WASM embedder. In your analysis, describe the structure of a WASM module, including its key components, and explain how the WASM embedder interacts with and executes the WASM module. Furthermore, discuss the role of WASI in enabling system interactions and the embedder's responsibility for isolation and security.

You are a cloud architect working for a company that provides image processing services. The company wants to migrate its existing on-premises image processing application to a serverless architecture in the public cloud to improve scalability, reduce costs, and simplify management. The application processes images uploaded by users, performs various transformations (e.g., resizing, cropping, filtering), and stores the processed images in a cloud storage bucket. The application needs to handle a large volume of image uploads, especially during peak hours. The company is considering using AWS Lambda, Google Cloud Functions, or Azure Functions for this project. They are also interested in exploring open-source serverless platforms like Knative or OpenFaaS as potential alternatives. The company has the following requirements:

• Scalability: The application must be able to automatically scale to handle fluctuating workloads without manual intervention.
• Cost-effectiveness: The solution should minimize costs by only paying for the resources consumed during image processing.
• Event-driven processing: The application should be triggered by image uploads to a cloud storage bucket.
• Integration: The solution should seamlessly integrate with other cloud services, such as cloud storage, databases, and message queues.
• Security: The application must be secure and protect sensitive data from unauthorized access.

Your task is to analyze the different serverless platforms and recommend the best option for deploying the image processing application. You should also discuss the security considerations and best practices for securing the application in a serverless environment.

1. Based on the requirements outlined in the scenario, evaluate AWS Lambda, Google Cloud Functions, and Azure Functions. Which platform would you recommend for deploying the image processing application, and why? Justify your recommendation by considering factors such as pricing, scalability, integration capabilities, and cold start times.

2. Describe the steps involved in deploying the image processing application using your chosen serverless platform. Include details on how to configure triggers, set up IAM roles and permissions, and monitor the application's performance.

3. Discuss the potential benefits and challenges of using open-source serverless platforms like Knative or OpenFaaS for this application. How do these platforms compare to the managed services offered by AWS, Google Cloud, and Azure?

4. What are the key security considerations for deploying a serverless image processing application? Outline best practices for securing the application and protecting sensitive data.