Feature/lab10

.github/pull_request_template.md

@@ -0,0 +1,13 @@
+# Goal 
+Describe the goal of the PR
+
+# Changes
+Describe the changes made to the codebase
+
+# Testing
+Describe the actions made to test the changes
+
+# Checklist
+- [ ] Clear title?
+- [ ] docs/README updated if needed?
+- [ ] No secrets/large temp files?

labs/submisssion10.md

@@ -0,0 +1,107 @@
+## Task 1
+
+### AWS
+Service name: AWS CodeArtifact
+
+Supported package formats:  Cargo, generic, Maven, npm, NuGet, PyPI, Ruby, Swift, generic
+
+Key features: high availability and durability, access control
+
+Integration with: AWS Identity and Access Management, CloudTrail, AWS KMS
+
+Pricing: Pay-as-you-go, separate bills for store/request/transfer
+
+### GCP
+Service name: Google artifact-registry
+
+Supported package formats: Docker, Maven, npm, Python, Apt, Yum, Kubeflow, Go, Debian, RPM, generic
+
+Key features: security, integration with Google's other services, remote and virtual artifact repositories, containers support
+
+Integration with: Google Kubernetes Engine, Cloud Run, Compute Engine, App Engine, Google Cloud CI/CD
+
+Pricing: Free plan available, Pay-as-you-go otherwise
+
+### Azure
+
+Service name: Azure Artifacts
+
+Supported package formats: Maven, npm, NuGet, Python, Cargo, Dotnet, Gradle, generic
+
+Key features: Integrated CI/CD, AI integration, availability during downtime
+
+Integration with: Azure Pipelines, Copilot
+
+Pricing: 30-day trial, Pay-as-you-go
+
+| Capability                       | **AWS CodeArtifact**                                          | **GCP Artifact Registry**                                     | **Azure Artifacts**                                   |
+| -------------------------------- | ------------------------------------------------------------- | ------------------------------------------------------------- | ----------------------------------------------------- |
+| **Containers support**           | No                                                            | Yes (native)                                                  | No                                                    |
+| **Public-registry upstreaming**  | Yes (npm, PyPI, Maven Central, NuGet)                         | Yes (via remote repos)                                        | Yes (npm, Maven, NuGet, PyPI)                         |
+| **Repository types**             | Private + upstream, but no true “virtual repos”               | Local, remote, virtual (aggregate)                            | Feeds with upstream sources                           |
+| **Storage model**                | Object-storage-backed, similar to S3                          | Region-scoped repositories                                    | Per-organization feed storage                         |
+| **Pricing**          | Free 2 GB + request-based billing                             | Region-dependent rates; heavier focus on network egress costs | First 2 GB per org free; mostly storage-based billing |
+| **Best suited for**              | Polyglot package repos inside AWS                             | Mixed containers + packages in one place                      | Teams using Azure DevOps                              |
+
+For a multi-cloud strategy GCP would be the best service among those, since it is more portable and repositories are region-scoped
+
+## Task 2
+
+### AWS
+
+Official services names: AWS Lambda
+
+Supported programming languages/runtimes: Node.js, Python, Java, Dotnet, Go, Ruby, custom are supported
+
+Execution models (event-driven, HTTP-triggered, etc.): event-driven, HTTP-triggered, scheduled
+
+Cold start performance characteristics: avg 100-2000ms
+
+Integration with other cloud services: basically all AWS ecosystem including S3
+
+Pricing model (per invocation, per execution time, etc.): Free plan available, Pay-per-use
+
+Maximum execution duration limits: up to 15min per instance
+
+Common use cases and architectures: S3 uploads and processing, microservices, light compute
+
+### GCP
+
+Official service name: Google Cloud Run
+
+Supported programming languages/runtimes: Node.js, Python, Go, Java, Dotnet, PHP, Ruby, generic
+
+Execution models (event-driven, HTTP-triggered, etc.): HTTP-triggered
+
+Cold start performance characteristics: avg 200-1500ms
+
+Integration with other cloud services: GCP ecosystem
+
+Pricing model (per invocation, per execution time, etc.): Free plan available, Pay-per-use
+
+Maximum execution duration limits: Plan-dependent, up to 60min per instance
+
+Common use cases and architectures: web backend, simple services
+
+### Azure
+
+Official service name: Azure Functions
+
+Supported programming languages/runtimes: Dotnet, Node.js, Python, Java, generic
+
+Execution models (event-driven, HTTP-triggered, etc.): HTTP-triggered, scheduled, event-driven
+
+Cold start performance characteristics: avg 200-3000ms
+
+Integration with other cloud services: Microsoft ecosystem
+
+Pricing model (per invocation, per execution time, etc.): Pay-per-use, "free" trial
+
+Maximum execution duration limits: shortest, 230 seconds
+
+Common use cases and architectures: Dotnet projects, enterprise microservices
+
+For a REST API backend I would use Google Cloud Run since it is concurrent and optimised for HTTP workloads (as per their documentation). Also idling is free.
+
+Advantages: provisioning and handling by the service, high availability, integration with respective ecosystems
+Disadvantages: cold starts, limited execution time