Despite already producing one of the most widely adopted and respected business intelligence platforms on the market, Microsoft unveiled Azure Synapse Analytics on Monday, a new system that joins data warehousing and data analytics in a single cloud service.
Though rolled out at the start of the tech giant’s annual user conference, Microsoft Ignite, Azure Synapse Analytics is in preview and not yet available on the general market.
Rather than replace any existing BI products, Azure Synapse Analytics will integrate with both Power BI and Azure Machine Learning, providing the data management that fosters analytics and data science, according to Microsoft. The result will be an improved ability for an organization to both glean insight from its data and apply machine learning to all of its intelligent applications, the vendor said.
“I think we’re seeing the next wave of innovation here, where hyperscale vendors are looking to better unify their growing number of analytics, big data and AI-based services,” said Mike Leone, senior analyst at Enterprise Strategy Group, based in Milford, Mass.
Mike LeoneSenior analyst, Enterprise Strategy Group
Given the unique features of data warehouses and data lakes, many organizations use both. Data warehouses and data lakes don’t, however, work in concert with one another, and without the systems communicating with one another the decision-making process suffers.
Azure Synapse Analytics attempts to bring data warehouses and data lakes together in a single experience to then ready data for BI and machine learning.
In effect, Microsoft said, Azure Synapse Analytics is the next step in the evolution of Azure SQL Data Warehouse.
“It sounds like they have enhanced SQL Data Warehouse to support external tables — a function that lets SQL Data Warehouse forward a local query to a remote database or data lake — without users having to know the data is remote,” said Wayne Eckerson, president of Eckerson Group, based in Hingham, Mass. “To them it looks like the remote data is local to the SQL DW database.”
Leone added that interaction between an organization’s different systems is crucial as BI evolves.
“The idea of breaking down analytic walls is essential for organizations to reach the next stage of data-driven insight,” he said. “While 60% of organizations use a mix of structured and unstructured data for analytics purposes, 45% use structured and unstructured data independently of each other. The silos need breaking down.”
But while the new system is an advancement for Microsoft, the unified query aspect of Azure Synapse Analytics, giving a user one place to look across their structured and unstructured data, is not necessarily groundbreaking. Presto, for example, is a SQL query engine that enables big data queries. AWS Athena does as well, and Oracle has a query option for both structured and unstructured data.
Azure Synapse Analytics, however, streamlines what Microsoft is offering for data query.
“This is something a number of providers have supported for a long time,” Eckerson said. “SAP calls it Smart Data Access, which lets HANA support external queries and thus dynamic queries to SAP IQ or Hadoop — or other — for example.”
What may be more unique than the data query capability is the claim that Azure Synapse Analytics not only provides a unified query model but does so for both BI and analytics users and data science users employing machine learning.
“This will enable customers to seamlessly connect and analyze more data across the analytics pipeline,” Leone said. “And by simplifying the ability to embrace next-generation technologies like machine learning, organizations gain peace of mind knowing they’re on track to continue transforming their business well into the future.
When Azure Synapse Analytics is ultimately released for public consumption, according to Microsoft, users will be able to pull together data from multiple sources more quickly than they could previously and do so in a secure environment. Microsoft partners, meanwhile, will continue to have access to a partnership ecosystem that includes vendors such as Accenture, Attunity (now a subsidiary of Qlik), Databricks, Informatica and Talend.