GCP Digest Issue 4

13 July 2020
Back to homepage
The amount of news coming out from Google Cloud in the past two weeks was at a dizzying pace. Writing this newsletter helps me to come to terms with it, and hopefully for you too.

To me, the biggest highlight would be that Cloud SQL for PostgreSQL finally lands support for point-in-time recovery (without fanfare). Also, there's the release of the Google Cloud AI Adoption Framework to help your organization on your AI adoption journey. And Sheets is really getting some really intelligent and serious chops with Connected Sheets, Sheets Smart Fill, and Sheets Smart Cleanup. Not to mention Data QnA, which allows you to query BigQuery without writing SQL queries, but rather ask questions in English!

On my end, I just gave a talk tonight on the topic of Cloud Tasks in GDG Cloud KL's July meetup! Check out the YouTube livestream.

Happy reading!
- Jonathan

Cloud SQL for PostgreSQL supports PITR (finally)

I've been following this issue for years now, and finally GCP has decided to unveil point-in-time recovery for Cloud SQL for PostgreSQL a few days ago, albeit with little fanfare. In fact, the people who caught the news and posted in the issue wasn't even from the Cloud SQL team! How unenthusiastic Googlers are when it comes to launching something that's actually a (late but) pretty big deal.

Point-in-time recovery helps you recover an instance to a specific point in time. For example, if an error causes a loss of data, you can recover a database to its state before the error occurred. Cloud SQL for PostgreSQL was lacking this for years (well okay, only 2 years since GA). A point-in-time recovery always creates a new instance. See the product documentation

Google Cloud AI Adoption Framework

You have probably heard about the Google Cloud Adoption Framework, which guides you on your move to the cloud.

Now GCP has also introduced the Google Cloud AI Adoption Framework (PDF) to help answer questions such as the following:
  • “Which skills should we hire and how should we structure our teams?”
  • “What ML projects should we prioritise?”
  • “How do we implement responsible and explainable AI?” 
  • “How can we make data and ML assets discoverable, shareable, and reusable?”
  • “How can we utilize cloud-native services to scale?”
  • “How can we operationalize data processing and ML pipelines in production?”
The AI Adoption Framework builds a structure on four areas: people, process, technology, and data. The interplay between these areas highlights six themes that are critical for success: Lead, Learn, Access, Scale, Automate, and Secure. Read more in the blog post.

Bare Metal Solution available in more regions

Bare Metal Solution consists of all the infrastructure you need to run your specialized workload such as Oracle Database close to Google Cloud. The solution connects to all native Google Cloud services via a dedicated, low-latency and highly resilient interconnect. Google Cloud deploys Bare Metal Solution in a region extension with less than two millisecond latency to Google Cloud; in most cases the latency is measured to be sub-millisecond.

Bare Metal Solution is now available in five more regions: Ashburn, Virginia; Frankfurt; London; Los Angeles, California; and Sydney. By the end of this year four more regions are planned: Amsterdam, São Paulo, Singapore, and Tokyo.

Google Cloud VMware Engine now generally available

"Lift and Shift" is now even easier if you are using VMware on-premises. Google Cloud VMware Engine enables you to seamlessly migrate your existing VMware-based applications to Google Cloud without refactoring or rewriting them. VMware Engine is available in the us-east4 (Ashburn, Northern Virginia) & us-west2 (Los Angeles, California) regions, with more to come later.

Google Cloud VMware Engine is a first-party offering, fully owned, operated and supported by Google Cloud, that lets you seamlessly migrate to the cloud, without the cost or complexity of refactoring applications, and manage workloads consistently with your on-prem environment. Check out the blog post.

Analyze BigQuery data from the comfort of Sheets

GCP has announced the general availability of Connected Sheets, which provides the power and scale of a BigQuery data warehouse in the familiar context of Sheets.

Connected Sheets allows you to analyze billions of rows and petabytes of BigQuery data from the comfort of Google Sheets. You're not really pulling data into the spreadsheet, but rather it lives in the BigQuery database. Familiar Sheets tools like pivot tables, charts, and formulas can be applied to big data, reducing the dependency on specialized analysts.

Sheets Smart Fill and Sheets Smart Cleanup

Google Sheets is on a roll. Apart from Connected Sheets, "later this year" G Suite will be launching Sheets Smart Fill, which detects and learns patterns between columns to intelligently autocomplete data for you. It is similar to Smart Compose in Gmail, but applied to Sheets.

Another thing coming soon is Sheets Smart Cleanup, which will surface intelligent suggestions in the side panel when you import data into Sheets. It can help you identify and fix duplicate rows and number-formatting issues.

Seeing is believing, check out the blog post for GIFs of how the above work. Sheets Smart Fill and Smart Cleanup will be available to G Suite customers later this year.

New Compute Engine A2 VMs (private alpha)

We already have N1, N2, N2D, E2, C2, M1, and M2 machine types in GCP. Well guess what, there's now A2!

The Accelerator-Optimized VM (A2) family of VMs in GCP allows you to run up to 16 GPUs in a single VM, and they are based on the NVIDIA Ampere A100 Tensor Core GPU, the first in a public cloud. It allows you to run the most demanding workloads such as CUDA-enabled machine learning (ML) training and inference, and high performance computing (HPC). Check out the blog post.

Cloud-agnostic data governance: Collaboration with Collibra

Data governance requires you to show where data has been stored, and how it’s been used, to meet regulations. It also requires you to use access controls and other data governance tools helps ensure that only those who need to see certain data are able to. For example, BigQuery natively supports fine-grained access controls at the column level.

GCP is now partnering with Collibra to offer a cloud-agnostic and source-agnostic data governance solution. Collibra’s technology will directly interface with Google Cloud security primitives, allowing your data governance policies to be natively enforced as direct-column, table-security elements at the storage layer. It also serves as an independent control plane to provide visibility to data outside of Google Cloud. Check out the blog post

Data QnA: Natural language interface for BigQuery analytics

Data QnA (now in private alpha) helps enable your business users to get answers to their analytical queries through natural language questions, without burdening business intelligence (BI) teams. Data QnA is based on the Analyza system developed at Google Research.

For example, you can issue the following questions to Data QnA and get back a predicted interpretation and result:

  • "Top products by average price"
  • "How many orders for Organic Strawberries"
  • "Top states by count of order for Organic Strawberries"
In most enterprises, when business users need data, they request a dashboard or report from the BI team, and it can take days, or even weeks, for the already overloaded team to respond. Data QnA enables self-service analytics for business users on BigQuery data as well as federated data from Cloud Storage, Bigtable, Cloud SQL, or Google Drive. See the blog post.

Reduce cloud complexity with Active Assist

Active Assist is a portfolio of intelligent tools and capabilities to help actively assist you in managing complexity within your cloud operations. It leverages data, machine learning, automation, and intelligence, to bring “Google magic” to you, so you can enjoy a simpler, smarter cloud experience in your day-to-day operations. 

Active Assist helps you with three key activities:

  • Making proactive improvements to your cloud with smart recommendations
  • Preventing mistakes from happening in the first place by giving you better analysis
  • Helping you figure out why something went wrong by using intuitive troubleshooting tools
You can't start using Active Assist yet, but to test out new capabilities before they’re made publicly available, you have to fill out a form to join their Active Assist Trusted Tester Group. Not sure about the pricing as well. See the blog post

Beta? GA?

The list below is best-effort and not meant to be exhaustive.

Entered GA
Entered Beta
For more product updates, visit Google Cloud release notes



For way more, check out TWiGCP's past two issues.

See you next time!

Send me GCP Digest

One newsletter, every two weeks.

We care about the protection of your data. Read our Privacy Policy.