Lifetime Citizen Portal Access — AI Briefings, Alerts & Unlimited Follows
eWorld demonstrates serverless web apps with Google Cloud in hackathon presentation
Loading...
Summary
eWorld Enterprise Solutions demonstrated a serverless album‑library web app built on Google Cloud Platform, explaining cost, scaling and operations benefits, comparing AWS/GCP/Azure, and answering attendee questions about media playback, hackathon use cases and version control.
Kerry Ho, director at eWorld Enterprise Solutions, and colleagues demonstrated how serverless architecture can power web applications using Google Cloud Platform (GCP), showing an album‑library demo that handled create, read, update and delete operations with Firestore, Google Cloud Storage and Cloud Functions.
The presentation began with Ho describing eWorld as “agnostic to a lot of these cloud providers,” saying the firm implements solutions on Google, AWS and Azure for government and commercial clients. Ari Asenfeld defined serverless as “an event driven architecture where things happen when they're triggered by HTTP requests or database updates,” and said the model shifts infrastructure management to the cloud provider so developers can focus on code and pay only for usage.
Presenters highlighted practical advantages: automatic scaling to handle traffic spikes, reduced operational overhead and lower barriers to entry for small teams. Ho described a migration case in which replacing persistent relational hosting with a serverless Aurora (on AWS) produced “huge cuts in expenses” by avoiding always‑on database instances. The team noted typical serverless cost features such as generous free tiers—Ari said providers commonly offer “1,000,000 free requests per month”—and low per‑request charges beyond those quotas.
The session covered technical tradeoffs across providers. Asenfeld compared AWS Lambda, Google Cloud Functions and Azure Functions, noting differences in language support, integrations and execution limits (Lambda up to 15 minutes; Google Cloud Functions about 9 minutes for event triggers and 60 minutes for HTTP functions; Azure Functions 5 minutes on the default plan and 60 minutes on premium). The presenters also explained ‘cold starts’—initial invocation delays—and optimization approaches like AWS SnapStart for Java.
For the demo, developer Kaelin Hirokala showed a web UI that adds albums (title, release date, description, track count), uploads cover art to Cloud Storage and stores metadata in Firestore. Kaelin demonstrated editing metadata, confirming the UI updates Firestore, and deleting entries to show the full CRUD lifecycle. She explained that Cloud Functions host the REST endpoints that the front end calls and that runtimes such as Node.js or Python can be used for the backend logic.
During Q&A, an attendee asked whether the demo could play music; the presenters said yes, explaining it would typically use a client‑side media player that references audio files stored in Cloud Storage. The team confirmed serverless approaches are suitable for hackathon challenges that collect or display data, and emphasized that version control and deployment follow standard Git and CI/CD patterns: store source in a repository, tag or branch the desired version and run the pipeline to deploy.
Presenters invited follow‑up questions in the Slack channel and at their in‑person booth. The session closed with logistical reminders and offers of additional support.
