Snowflake Computing Inc., a San Mateo-based company, announced the availability of its data warehousing platform on Microsoft Azure.
The evolution of the public cloud has disrupted the traditional software industry. During the last decade, almost every mainstream enterprise software platform became available in the cloud. This paradigm shift created new opportunities for the independent software vendors to build born-in-the-cloud applications and platforms that challenge the incumbents.
Snowflake Computing is one of the first few companies to realize the potential of building a native, cloud-based data warehouse. Founded in 2012, this company exploited the capabilities of the cloud to design a SQL data warehouse from the ground up. Snowflake’s data platform turned all the knobs available in the cloud to extract the best performance. Within a short span, it started to pose a threat to traditional data warehousing companies.
Given its popularity and adoption, Snowflake chose AWS as its chosen cloud for building the data warehouse platform. With the explosion of Big Data, Snowflake rode the wave of data analytics.
Snowflake followed the best practices of building applications for the public cloud. It leverages object storage to ingest and store semi-structured data based on formats such as JSON, AVRO, and XML. Customers can use familiar ANSI SQL to query the data. Developers can use standard JDBC layer to connect applications to the warehouse. The heavy lifting involving data processing and querying is delegated to an elastic compute layer that can dynamically grow and shrink. An orchestration layer that acts as the control plane manages the entire stack. This decoupled architecture of the data plane and control plane delivers maximum flexibility and efficiency to customers.
Snowflake is now bringing its cloud-ready data warehouse to Microsoft Azure. It mapped all the layers of its stack running on AWS to Azure. The data warehousing platform uses Azure Storage and Blobs to store raw data. The unstructured and semi-structured data can also be ingested via Azure Data Lake. The compute layer takes advantage of the beefy virtual machines on Azure that deliver the horsepower required for processing. Microsoft customers can use Power BI Desktop for visualizing the results from Snowflake.
Bob Muglia, CEO of Snowflake, is an industry veteran with decades of experience in building enterprise platforms. At Microsoft, he was instrumental in transforming SQL Server into an enterprise database. As an SVP of Servers and Tools, Muglia was responsible for building the application servers and developers tools based on Microsoft .NET. He joined Juniper Networks as an EVP before becoming the CEO at Snowflake.
Given Muglia’s affinity to Microsoft, it’s not surprising to see Snowflake finding its way to Azure. Technically, Azure has all the building blocks and core services that match AWS making it possible to map the architecture.
With its move to Microsoft Azure, Snowflake becomes one of the few multi-cloud data warehouses in the market. Apart from competing with traditional, on-premises data warehouse vendors, it’s already emerging as a viable alternative to managed data warehouses such as Google BigQuery, Amazon Redshift and Azure SQL Data Warehouse.
