The overwhelmed person’s guide to Google Cloud: week of February 1

Google Cloud Platform · February 6

The content in this blog post was originally published last week as a members-only email to the Google Cloud Innovators community. To get this content directly in your inbox (not to mention lots of other benefits), sign up to be an Innovator today.

New and shiny

Three new things to know this week

Serverless … but stateful? One hallmark of serverless computing is that it’s ephemeral and instances can come and go based on demand. What if your workload needs access to durable storage volumes? Now in preview, mount Cloud Storage buckets to Cloud Run volumes.
AlloyDB is getting better almost every day. This powerful PostgreSQL-as-a-service now supports PostgreSQL 15 for new or existing clusters. You can also restore cluster backups across projects. And we just shipped a series of language-based connectors which make it easier for Java, Go, and Python apps to connect to their database.
Trigger integration workflows via webhooks. Have you tried the Application Integration service? It offers a powerful way to create declarative integrations between systems. Now you can kick off these integrations whenever you invoke a webhook. Very handy!

Watch this

Google Cloud jumpstart solution

An empty cloud project is a lonely cloud project. Jump Start Solutions offer an outstanding way to quickly deploy a reference app and explore how cloud services work together. This playlist points to a handful of walkthrough videos that explain some of the most popular ones.

Community cuts

Every week I round up some of my favorite links from builders around the Google Cloud-iverse. Want to see your blog or video in the next issue? Drop me a line!

Good developers clean up after themselves. Rodrigo shows us how to create Terraform that sets up a Vertex AI Notebook environment that uses the auto-shutdown feature.
There’s no one “right” way to do data integration. Many tools can play a part when your goal is to move data around. Mazlum uses Cloud Workflows here, but also explains the role that products like Cloud Composer play.
Keep showing us how you Duet. You all have great ideas about where AI assistance can help. Kondala uses this article to explain how he created a GKE cluster and deployed a workload to it.
What happens when data processing jobs fail halfway through? The risk with long-running data processing tasks is that something may go wrong before the task completes. Nikhil offers an educational post on how Google Cloud Dataproc handles faults.

Learn and grow

Three ways to build your cloud muscles this week

I like that this post shows us the powerful combination of BigQuery with the popular framework LangChain. Follow along here to build a data loading solution with LLMs.
Using Cloud Deploy to get your models to Vertex AI? I like it, and Ivan produced a useful walkthrough that helps you uplevel your ML game.
Will you be processing pictures of fruit in your production system? Probably not, but many generative AI demonstrations are about sparking your creativity. This post shows you how to analyze images by using Remote Functions to call Gemini endpoints from BigQuery.