Your Data Cloud Questions - Answered!

[mesnaider]

Missed the Live Data Cloud Q&A session at Google Cloud Next? No worries! We’ve transcribed both the questions and the answers given by VP & GM of Database & Analytics, Gerrit Kazmaier, Director of Product Marketing for Data Cloud, Dain Hansen, as well as the live demo speakers, Leigha Jarett and Nikita Namjoshi. We’ve also added bonus questions and answers that the moderators didn’t have time for during the live session.

If you have any further questions about all things data and Google Cloud, please comment on this article and our team (or someone from the community) can help you out!

Question 1: What is the availability of Earth Engine on Google Cloud? And - how does it work with BigQuery GIS?

Answer: Google Earth Engine is in preview. Go to this link to fill out a form and our team will get back to you to set you up with an account!

In Earth Engine, you have a data lake of satellite imagery, and you can do geospatial analysis on that imagery data. Those results can be brought into BigQuery to match them up with all of the data you already have on BigQuery (or using BigQuery ML).

Learn more about Earth Engine on Google Cloud here.

Question 2: What are Google’s plans around data management? Does Google have plans to expand their multi-cloud approach to other data management services?

Answer: For the majority of companies, multi-cloud is their reality. Hence, a modern data silo is a multi-cloud data silo. With BigQuery Omni (announced at Next ‘21 Day 1) we wanted to bring simplicity to multi-cloud - meaning you are now able to do cross-cloud analytics… I think ~90% of companies that we’ve surveyed actually come back to us saying that they’re in a multi-cloud landscape today. We definitely aim to follow our users and meet them where they are, so you can expect much more from us in this direction moving forward.

Learn about BigQuery Omni here.

Question 3: Can you visualize the semantic model that you’re creating in Looker to connect all of your different data assets?

Answer: Yes! Earlier this year, Looker launched the LookML diagram (available in Looker’s marketplace). This means, you can create a visualization of your data model directly in Looker, whether your data is in BigQuery, Spanner, or other places that are connected to Looker. 

Learn more about LookML here.

Question 4: When should I use Cloud SQL and when should I use Cloud Spanner?

Answer: Cloud SQL is a managed service where you can decide whether you want PostgreSQL or MySQL, and it’s for customers who want to leverage perfect compatibility with the PostgreSQL or MySQL ecosystem. It’s highly reliable and high performing.

Spanner sits on the other side of the spectrum - it’s a cloud native database that combines a unique set of abilities - it has 99.999% availability - that kind of availability is unheard of! It has global consistency and it’s horizontally scaling. If you’re building massively-scaling global applications or are all about scale and having strong consistency, Spanner is the way to go. Plus, the good news from today is that we just announced the PostgreSQL interface for Spanner. This means you don’t need to relearn anything or learn another tool set in order to work with it. Instead, you can seamlessly apply your PostgreSQL skills and go cloud native with Cloud Spanner.

Learn more about Cloud SQL here.

Learn more about Cloud Spanner here.

Learn more about the PostgreSQL interface for Spanner here.

Question 5: How does the Looker partnership with Tableau help our Looker customers in particular?

Answer: It’s all about simplicity and openness. This announcement means that if you’re a Tableau customer, you’ll be able to use Looker’s semantic models built in LookML for trusted data access, for data governance, for high performing database access, etc… If you’re a Looker customer, it means that you can add Tableau’s self-service data visualization right onto your Looker scenario.

So the bottom line is that it makes working with data simpler for everyone. It helps our customers who are trying to find the right balance between self-service and flexibility on one side, and strong governance and semantics on another. It really brings the best of two worlds together. It’s a reflection of the open strategy of the data cloud - it brings more community, more customers, more use cases, and more capabilities that we bring as part of one ecosystem.

Learn more about the new Looker integrations here.

Question 6: In the Data Cloud Live Demo you showed an integration with Vertex AI. How does this work with BigQuery ML?

Answer: For context, BigQuery ML allows you to build and train machine learning models with just a few lines of SQL. So in the demo we showed how you can use Vertex AI Workbench (announced at Next!) to have access to essentially the BigQuery console, but from a notebook interface. So you could actually create and train ML models in SQL and do all of this from the notebook interface and still have that connection to BigQuery and still use whatever popular data you have stored in BigQuery.

Learn more about Vertex AI Workbench here.

Question 7: How is Looker enabling new sets of API-driven experiences? What should we consider when we integrate Looker with other tools?

Answer: First off, it’s important to note that Looker is going to be a key element of any data landscape. Looker is giving you two amazing things: 

1) A model semantic layer where you can describe your data landscape with very rich annotations. This semantic layer is connected directly to the database system. For instance, when it runs on BigQuery, it completely delegates all of the queries all into BigQuery so it delivers stellar performance. 

2) Looker as a whole has a pretty amazing API-first concept. It’s so beyond BI (Business Intelligence) that it allows you to build data-rich applications. It allows you to run your custom experiences, your websites, your internal applications and make them really data rich. 

And last but not least, we’re also integrating more and more products with Looker’s semantic model. For instance, we’re adding Connected Sheets as a way to analyze and discover data described in LookML. Check it out - it’s going to make a big difference in your data landscape as well.

Learn more about the new Looker integrations here.

Question 8: How does Earth Engine connect over 50 petabytes of data to enable various ways to process data with various tools? 

Answer: First of all, it starts with making sure that there is data. And with Earth Engine you get an amazing data lake of satellite imagery and GIS data available to you. The second piece is that you need to have a powerful processing framework, and with the integration to BigQuery, you can use the most scaling, and fastest warehousing system on this planet to analyze this data alongside all the rest of your data and of your business data. Last but not least, it’s about complete openness. The announcements that we have shared today with LookML allows you to connect Tableau, Connected Sheets, Looker visualizations, etc. all on top of the right tool, ultimately ensuring that you’re using the visualization tool that you love the most.

Learn more about Earth Engine on Google Cloud here.

----------------

Wait, there’s more! Two additional questions came in about the new PostgreSQL interface for Cloud Spanner that unfortunately Gerrit Kazmaier and Dain Hansen didn’t have a chance to answer live. The Google Cloud data teams were happy to provide those questions and answers here below for everyone in the community!

Bonus Question 1: How should customers choose between the existing Google SQL interface and the new PostgreSQL interface for Cloud Spanner?

Answer: We aim to provide the same set of capabilities in both Spanner interface options: Google Standard SQL and PostgreSQL. Users should choose the interface that best aligns with the ecosystem in which they work. The core of Google Standard SQL is common across BigQuery and Spanner, allowing teams to share queries and concepts.  Existing Spanner customers can be assured that Google is fully committed to supporting and improving Google SQL in Spanner. Customers standardizing on PostgreSQL now have an option to take advantage of Spanner’s differentiators using the tools and skills they already have. 

Bonus Question 2: How can you sign up for Cloud Spanner’s PostgreSQL interface preview?

Answer: The PostgreSQL interface is a new feature of Spanner. It's available as a preview at no additional cost. You can sign up by completing the form at https://goo.gle/PostgreSQL-interface. We will evaluate the submissions and get back to you. 

Thanks for reading along! Curious to see the Data Cloud spotlight address from Gerrit Kazmaier and Stephanie Wong (@swongra)? Check it out here. And see all of Google’s data platform announcements here.

If you have any further questions about all things data and Google Cloud, please comment on this article and our team (or someone from the community) can help you out!