We are now part of Eviden, discover more...

Powering Enterprise AI Projects with BigQuery

Google Cloud’s BigQuery has come a long way since it launched in 2010, and even since machine learning (ML) was first added to the serverless, scalable cloud data warehouse last year. Google CEO Sundar Pichai publicly introduced the company’s Artificial Intelligence (AI) – first strategy in 2017 and focused on “making AI ‘simple, fast and useful’ for enterprises” in the months following, according to CMSWire.

The company’s investments have followed suit. Google Cloud has been continuously updating BigQuery, for one. In one example, cost optimization upgrades are giving enterprises more options than ever to minimize cost while ensuring that they have optimal functionality. Specifically, users can balance fixed price and on-demand options on a project-level basis. Just last month, Google also added scripting to the BigQuery platform, enabling developers to run queries in a sequence or conduct multi-step tasks with control flow within the platform. And with new stored procedures capabilities in BigQuery, users can also save those scripts to run again in the future.

As for BigQuery ML, in particular, the team recently enhanced its original logistical and linear regression models to include K-means segmentation. Now, users can tie ML directly into ETL pipelines, which is a big jump forward when it comes to ease of implementation and productionalization of certain use cases.

ML & AI with BigQuery

Since BigQuery ML was launched, users now have the ability to create ML models using standard SQL queries. BigQuery ML is a set of simple SQL language extensions that enables users to utilize popular ML capabilities without advanced ML skills. The platform sets smart defaults automatically to take care of data transformation, leading to a seamless and easy-to-use experience. It supports these ML models:

  • Linear regression — used for predicting a numerical value
  • Binary logistic regression — used for predicting one of two classes (such as identifying whether an email is spam)
  • Multiclass logistic regression for classification — used to predict more than two classes such as whether an input is “low-value,” “medium-value,” or “high-value”

With BigQuery ML, data analysts no longer need to move data between environments; the tool has significantly reduced the friction associated with producing ML models. In addition, BigQuery ML democratizes ML – no python or other coding experience required. Learn more here.

How Does BigQuery Compare To Other Data Warehouses?

Other querying services do exist for AI and ML applications, of course. But BigQuery offers a few clear differentiators: 

  • BigQuery is native to GCP and fully serverless, which makes usage straightforward and billing simple. With competitors that function as a third party on other clouds, server management overhead with regard to the amounts of compute needed at any given time can be limiting. In this case, some degree of traditional capacity planning needs to be involved in order to handle large usage spikes.
  • Google has led the way in providing more places for ML to occur. BigQuery ML provides one of the simplest ways to develop, test, and operationalize ML results by simple scheduling queries and sharing the results. In addition, unique use cases for BigQuery ML exist around leveraging ML within standard SQL reporting tools.
  • BigQuery simplifies data storage compared to other databases with fewer data types. The platform stores data in files rather than storage blocks. 
  • As a leading cloud-native database, BigQuery lends database users more power and flexibility than they have ever had in the past. The platform is making waves when it comes to analytical processing that advances business capabilities and simplifies IT processes for many companies.

Find Out More

To read more about BigQuery, download “A Guide to Google BigQuery Inner Workings” here. We must add, too, that Google Cloud’s AI tools for enterprises go far beyond BigQuery. Cloud AutoML allows developers to train customer ML models. AI Hub is data scientists’ one-stop-shop for ML content, and those are just a couple of examples. 

Maven Wave helps enterprises, like ADARA, migrate to and manage BigQuery. In ADARA’s case, the company observed a 30% reduction in operational cost while enjoying BigQuery’s nearly limitless ability to scale automatically. To learn how BigQuery can help your enterprise harness AI, contact us.

About the Author

Kylie McKee
Kylie McKee is a Content Marketing Strategist at Maven Wave with more than eight years of tech industry experience and five years of content marketing experience. Prior to joining the Maven Wave team, Kylie worked as a Content Marketing Specialist for WebPT, Inc. and earned an Associate in Applied Science in Motion Picture, Television, and New Media Production with a CCL in Screenwriting from Scottsdale Community College.
January 31st, 2020

Get the latest industry news and insights delivered straight to your inbox.

Sign up for our Newsletter