Google Cloud Platform Technology Nuggets — July 1–15, 2024 Edition

Welcome to the July 1–15, 2024 edition of Google Cloud Technology Nuggets.

Please feel free to give feedback on this issue and share the subscription form with your peers.

Containers and Kubernetes

If you are responsible for keep your GKE cluster up and running, chances are that there are multiple times that you might have hit networking issues. This blog post goes into common networking issues, grouped by categories and then lays out a step by step strategy on you can go about troubleshooting them. This can be an excellent playbook that you could start with your organization.

Identity and Security

Identity and Access Management, is the cornerstone of your security when it comes to your applications, cloud resources, users and who can do what operation on it. But this can be a difficult path to navigate with lots of terms and any misunderstanding, leave your environment susceptible to unpleasant outcomes. Check out this guide that dives into IAM, highlights the key terms, foundational building blocks and how you can navigate it. Definitely a blog post to bookmark.

Sensitive Data Protection (SDP) is a fully managed service designed to help you discover, classify, and protect your valuable data assets with ease. Cloud Storage is now the latest service to be supported by SDP in addition to BigQuery, Big Lake and Cloud SQL. Take a look at the blog post that highlights the scope at which you can apply this service, various features of the service and how you can integrate this into your security posture.

Networking

Here is an interesting blog post that first talks about Cross-Cloud Network, that enables an open, secure, and optimized way to operate your enterprise network between different cloud environments. It then goes on to highlight the recently announced partnership betwee Google Cloud and Oracle Cloud, which adds to the Cross-Cloud Network offering by highlighting how it enables customers’ Virtual Private Clouds (VPCs) in Google Cloud to connect to their Virtual Cloud Networks (VCNs) in Oracle Cloud Infrastructure (OCI). Check it out to learn the first regions that the service will be supported and how it works.

Machine Learning

You’ve been busy building out GenAI applications and its time to step back and check if the LLMs are giving you the desired results. In essence, how do you evaluate a LLM? Is there a framework to do so? Is there a service from Google Cloud that helps you to do that in a streamlined way. Enter the Vertex AI Evaluation Service and check out this deep dive that provides instructions on how you can use it today.

Most organizations find that eventually as they build out their bots (agents) for customer service, they need a platform that provides a lot more than just developing/deploying bots. You need the platform to integrate with existing telephony systems, a developer / deployment platform for your bots, analytics from the bot usage to understand its usage and scope of improvements, a service to help out human agents complete the last mile of the conversation and more. Google Cloud has a Contact Center AI Platform that has been there for a while and worth a look at, if you haven’t. Check out this blog post that highlights what CCAI is, the different components and how you can get started with it.

Storage and Databases

Google Cloud has been highlighting AlloyDB, a fully managed, PostgreSQL-compatible database service as a powerful option that in addition to PostgreSQL compatibility, also provides a solid price-performance return, Gen AI features and Google scale and availability. The challenge has been to be able to try it out with as less friction as possible. Google Cloud has now made it impossible to ignore with a free trial cluster that is available for you to take AlloyDB for a spin. The trial cluster comes with 8 vCPU and 64 GB RAM primary instance and 1 TB of regional storage. Check out the blog post for details, the link to get started with the trial cluster, quick start tutorials and more.

Google Cloud SQL has a new automated upgrade tool. It is available for major versions and Enterprise Plus, specifically available for MySQL 5.7 to 8.0.31, and PostgreSQL 9.6, 10, and 11 to PostgreSQL 14 and 15. The tool provides automated pre-upgrade assessments and reports, automated scripts to resolve issues and an in-place major version and Enterprise Plus upgrades. Check out the blog post on how you can use it today.

Data Analytics

BigQuery Canvas has a clear goal and that is to simplify the entire data analysis journey — from data discovery and preparation to analysis, visualization, and collaboration — all in one place, all within BigQuery. It depends heavily on Gemini in BigQuery, which allows you to use natural language queries throughout the different tasks and stages. If this is of interest, do check out this detailed blog post, which takes you through the entire process of using BigQuery canvas. The blog post uses the github_repos dataset, which contains over 3TB of activity from 3M+ open-source repositories. The blog post shows how you can answer queries on pertaining to Github operations performed on these repositories.

Continuing on with BigQuery Canvas, it is important to understand that behind the scenes it is a Gemini model that is powering it. This means that you have to provide effective prompts to the Gemini model, which in turn can then help you produce accurate queries using the NL2SQL and NL2Chart. Check out the blog post that should be your go to prompt guide for working with BigQuery Canvas.

Do you use BigQuery Analytics Hub to share datasets across organizations? What if you wanted to share real-time streaming updates to this data with various organizations? Enter public preview of Pub/Sub topics sharing in Analytics Hub. Check out how the feature is designed and can be put to use today.

Developers and Practitioners

Working with API Management tool Apigee and need to boost up the developer productivity while working with it? You should look at the Cloud Code plugin for API Management that has now got Code Assist support in Preview too. Check out the blog post for more details.

If you are building Go applications, it would help to learn more about Profile Guided Optimization (PGO). PGO is a process by which you could tell the compiler exactly how your code is used at runtime. This can have a good impact on improving the CPU utilization, thereby directly impacting costs too. As per the blog post, “you can collect a profile of your application in production and then use this profile on the next build, the compiler can make better-informed decisions like more aggressively optimizing the most frequently used functions, or more accurately selecting common cases inside a function.” Check out more on this topic and how you can use this in your application.

Learn Google Cloud

The topic that we will touch upon in this edition of the newsletter is not necessarily a Google Cloud 101 topic but something that is valuable to developers/operators or if you’d like, any user of Google Cloud. And that is the impact of GenAI on these roles, especially that of improving the developer experience. Check out this article that highlights what it is calling the “full-stack impact of AI”, where various disciplines like application development, DevOps, site reliability, machine learning, data, security, QA, and software architecture, are all impacted with the integration of GenAI capabilities. The article looks at Build, Operate and Secure aspects of software development and deployment and how various Google Cloud GenAI services are helping accelerate and improve productivity.

Stay in Touch

Have questions, comments, or other feedback on this newsletter? Please send Feedback.

If any of your peers are interested in receiving this newsletter, send them the Subscribe link.

Want to keep tabs on new Google Cloud product announcements? We have a handy page that you should bookmark → What’s new with Google Cloud.