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

Romin Irani

Published in

Google Cloud - Community

9 min readJun 17, 2024

Welcome to the June 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.

Get Learning today !

Have you heard of Google Cloud Innovators? It’s a community of Google Cloud enthusiasts, where you get access to news and latest events on Google Cloud. This program is free to join and open to everyone.

The program is now doubly attractive to anyone who is invested in the Google Cloud ecosystem via the announcement that all Google Cloud Innovators will get 35 no-cost learning credits every month! Yes, that’s correct every month that you can use to practice hands-on labs, access on-demand courses and more. These credits get renewed every month.

Join the Innovators program at no cost today!

Infrastructure

In a previous installment of this newsletter, we covered 5 ways that you could save costs while using Compute Engine in Google Cloud. The follow up to that highlights another 5 ways that you could save costs. Some of the recommendations include using Spot VMs, automate the process of turning machines on and off, letting go of external IP Addresses and more. Check out Part 2 of the series.

The N4 and C4 machine series in the Compute Engine VM portfolio were announced at Cloud Next 2024. The N4 machine series is generally available today and the C4 machine series is available now in preview for Compute Engine and Google Kubernetes Engine (GKE). Check out the blog post that provides more details on these series.

Partners

Developer Experience has a direct impact on creating modern applications, that in turn drives positive business outcomes. The SDLC (Software Development Life Cycle) is complex, filled with multiple tools, areas of possible attacks, plus we have regular context switching and disruption of the developer workflow. The recently announced Gitlab integration, a partnership between Gitlab and Google, highlights what a comprehensive solution looks like that addresses development efficiency, enhances security and overall improve software delivery. Check it out.

Identity and Security

Encryption and decryption between BigQuery and Sensitive Data Protection is now available. This makes it easy to protect and share sensitive data inside of BigQuery. Check out the blog post for more details.

Security Talks 2024 is scheduled to happen on June 26, 2024. You can RSVP here at the official event site and check out the agenda too. It comes as no surprise that there are several sessions that discuss the intersection of AI and Security.

A recent critical security vulnerability (CVE-2024–3094) discovered in the XZ Utils compression software affected sshd, a tool for secure remote access. This vulnerability could allow attackers to execute remote code on servers without requiring authentication. The blog post states that “While Google Cloud Compute Engine customers who rely on our supported public images are not affected by this vulnerability, those using custom images should refer to our security bulletin for remediation guidance.” The blog post also goes into recommending Identity Aware Proxy (IAP) and Security Command Center (SCC) , two tools that can help to strengthen your security posture and implement best practices for internet-exposed cloud resources.

Containers and Kubernetes

When I heard the word GKE Compliance, I was not sure what it meant? Is it is set of standards defined and accepted by the community? Given the type of workloads, does it apply to some of them? From the blog post, it is a streamlined way to assess your GKE clusters and workloads against industry-standards, benchmarks and control frameworks, including:

CIS Benchmark for GKE: The gold standard for secure GKE configurations
Pod Security Standards (PSS), which offer both baseline and restricted profiles to protect your workloads

The good news is that GKE Compliance is built into GKE and fully managed by Google. The GKE Compliance dashboard gives you centralized compliance reporting that is updated every 30 minutes, giving you a clear view of your compliance posture for your fleet of clusters.

Machine Learning

Google has been announced a Leader in The 2024 Forrester Wave™: AI Foundation Models for Language, Q2 2024, receiving the highest scores of all vendors evaluated in the Current Offering and Strategy categories. The full report can be accessed here: The Forrester Wave™: AI Foundation Models for Language, Q2 2024.

Gemini Integration with Google Workspaces and Google Cloud is likely to see some innovation solutions/integrations being built out. How do you get started on the same with foundational code that you could use to potentially implement some ideas that you might have for integrating the two services. Check out this blog post that show how to integrate Gemini and Google Sheets with BigQuery.

How do you monitor models deployed on any serving infrastructure (even outside of Vertex AI, e.g. Google Kubernetes Engine, Cloud Run, Google Compute Engine and more)? As the blog post states, that’s what the new Vertex AI Model Monitoring is for.

Finally, looking to implement RAG using BigQuery and LangChain? This practical guide gives you a step by step process to implementing it.

Storage and Databases

Cloud SQL for PostgreSQL now supports PostgreSQL 16. This new version brings in improvements in observability, performance, replication and more. Check out the blog post for more details.

With the recently announced LangChain packages for Cloud SQL for PostgreSQL, developing your AI app can receive a significant efficiency boost. Check out a blog post that highlights these LangChain integrations for using Cloud SQL for PostgreSQL as a vector database and if you’d like, there is an accompanying notebook too.

While on the topic of Cloud SQL, the rapid pace of integrating AI features into the databases continues. Several new features for Cloud SQL for MySQL, available in Preview, have been announced that help companies power their database and applications with AI. You can now use Vector Search to build generative AI applications and integrate with MySQL with the support for Vector Embeddings in Cloud SQL for MySQL. Additionally, with Gemini integration you can get advise on Indexing, debug performance issues in queries and more. Check out the blog post for these interesting features.

Running your enterprise workloads seamlessly across Google Cloud and Oracle Cloud Infrastructure (OCI) would have been something that one could not have imagined a while back. Well, this is going to be a reality with the recent announcement. Central to this concept is what is termed as Oracle Database@Google Cloud, where Oracle will directly host, operate and manage Oracle database services natively within and from the Google Cloud data centers, beginning with regional footprints in North America and Europe, and plans to rapidly expand globally. Read the blog post for more details and the upcoming integration.

Data Analytics

We are increasingly seeing the integration of Generative AI features across various Google Cloud services and BigQuery has definitely seen tons of features that make using Gen AI models inside of BigQuery a breeze. A popular use case has been that of Sentiment Analysis and this is an excellent blog post that dives into showing you how you can use Gemini models in BigQuery for that. Right from creating the data to extracting the themes and visualizing it, the blog post has it all.

Looking to generate synthetic data for facilitating the training of machine learning (ML) models or the evaluation of mathematical models? BigQuery DataFrames allows you to generate artificial data at scale. It is open-source python package providing pandas-like DataFrame and scikit-learn-like ML library for big data. It utilizes BigQuery and the rest of Google Cloud as the storage and compute platform under the hood. Check out the blog post for a step by step execution on how you can do that. There is accompanying notebook too.

Continuing with BigQuery DataFrames, here is another blog post that demonstrates how you can combine BigQuery DataFrames with the visualization capabilities of CARTO to build out complex mapping applications. The solution utilizes pydeck-CARTO, a Python library that renders deck.gl maps in a Jupyter notebook.

The name DataSageGen could be interpreted in multiple ways but as per the blog post, DataSageGen is a chatbot designed to be a personal guide to access and process information from a vast array of sources, including:

Data and AI product documentation
Blog posts and white papers
Community knowledge
Product and event announcements

How do you build such a bot? The blog post gives a step by step guide to doing so. There is a Github repository too.

BigQuery news has been heavy in this edition of the newsletter and to round it off, BigQuery User Defined Functions (UDFs) from the bigquery-utils repo are now available in all BigQuery regions.

Developers and Practitioners

Cloud Run, in my opinion is one of those services on Google Cloud that is so well suited to developer requirements of having a runtime that can cost applications easily and which can scale via its serverless nature. With the increase in the number of developers trying out AI services, Cloud Run is still a service that is possibly their best bet to get their AI applications out there to the world. The usual questions once you move your application to production includes whether the application will scale? Will we have enough monitoring features? How about revisions to A/B test out some scenarios? Check out this blog post that explains these points beautiful and a quick run through of taking a sample code from the Vertex AI studio and packaging it as a Flask application to deploy on Cloud Run. Probably the fastest way, eh ?

Earlier in this newsletter, we mentioned about a new post that highlights 5 more ways to save on Compute Engine costs. It looks like that “5 more” series seems to have caught on with everyone and we have a post titled “5 more myths about Platform Engineering”. This series is an excellent way to understand Platform Engineering and bust the myths one at a time. This part grounds everyone in their expectations on what Platform Engineering can solve, reduce dramatically and more.

DevOps and SRE

Looking to understand and learn more about building reliable systems but feeling a bit lost with the amount of material out there. Google Cloud has curated a list of resources to keep the focus and help you in this journey. Check out the blog post that provides research papers, blogs and books that you could look up to get up to speed on systems engineering.

Learn Google Cloud

Instead of a specific course or material to read up in this edition, this is a reminder again to join the Cloud Innovators program, that will provide you 35 credits every month to use towards courses and hands-on labs.