Four data analytics developments worth watching from the Snowflake Summit
June 20, 2021
For those who work with cloud data, the annual Snowflake Summit offers an opportunity to network and take the temperature of the industry.
Judging by the scale of the conference and new developments in the works, the data analytics industry remains in a state of expansion.
The coronavirus pandemic led to a virtual 2020 Snowflake Summit of a half-day conference with 40 sessions. Given COVID precautions remain in place, this year’s virtual event encompassed two days and more than 70 presentations.
No conference has yet found a way to replicate the lively, in-person experience of meeting exhibitors along the midway. But in terms of content, the event ran as smoothly as a TV show. Given the company recently became fully distributed after being based in Silicon Valley, Snowflake is no stranger to virtual collaboration.
For those tuning in, Snowflake Summit offered insights to where the company—and the industry—is headed in the future. Below, we recount four key takeaways from this year’s conference.
1. Snowpark opens up a playground for data developers
Having reached Elite status within the Snowflake Partner Network (SPN), DAS42 has long recognized Snowflake as a top choice for a data cloud provider. And with the announcement of a new development tool, the company is also highlighting its capabilities for clients beyond providing warehouse and analytics services.
Revealed last fall, Snowflake’s developer experience Snowpark follows the ongoing industry trend of centralizing data functions under a single platform. Though the platform has focused on using SQL for its management and queries, Snowpark will enable customers to run Python and Java-based programs directly in Snowflake.
As a result, your team’s data engineers, data scientists, and analysts can apply familiar programming concepts and coding languages to your data warehouse. Rather than needing an outside framework like Spark or Dask to execute processes, Snowflake will streamline development by offering its own programming capabilities.
In June, Snowpark customers on AWS gained support for Scala, which can be run on Java Virtual Machine (JVM). According to Snowflake representatives, support for Java, Python, and their related libraries and routines will roll out later this year.
2. Snowflake extends java support to user-defined-functions (udf)
As Snowpark introduces support for Java, the platform will also add new user-defined-function capabilities. Constituting a further expansion of Snowflake’s capabilities, the addition of UDFs will allow your teams to bring their custom code, third-party libraries, and business logic directly to where your data is stored.
These Java-based tasks can perform ETL transformations or engineering tasks to facilitate data analytics workflows. Though still in beta — or what Snowflake calls private preview — the functionality will soon be available to the public.
To some analysts, Snowflake’s addition of greater Java support sets the firm on the path toward competing with big players in the data cloud industry. To us, it opens up new possibilities for better performance and less complexity in your data stack.
3. Increased support for unstructured data in Snowflake
If your organization generates a lot of unstructured data, and exciting development from the Summit should appeal to you. “Unstructured” refers to any information that isn’t arranged in a preset database format, which could include images, video, or document records. Up to now, Snowflake customers who wanted to store unstructured data needed to incorporate a data lake into their data stack.
At Snowflake Summit, the company announced customers will now be able to use the platform to store, process, and share unstructured data. For companies that produce a lot of unstructured data, such as the gaming industry, Snowflake is introducing new possibilities for eliminating data silos.
Plus, organizations gain new possibilities for their analytics efforts through Snowflake’s increased compatibility with Java and Python-coded applications. With unstructured data open to analysis, your organization can access insights such as sentiment analysis reports from text or voice messages.
Though still in private preview, support for unstructured data should be coming to Snowflake soon.
4. Storage is growing more efficient and less expensive
As time goes on, all technology grows faster and more efficient. Similarly, as Snowflake manages its data format, the company continues to optimize its capabilities for the benefit of customers.
The advance of the cloud has offset costs of storage once tied to physical capacity and maintenance. But every company still has to pay for whatever space it uses. To offset those expenses, data warehouse companies apply data compression algorithms to their customers’ data.
At Snowflake Summit, the company announced its compression capabilities have improved, which further reduced costs associated with warehouse requirements. The improvements have already been rolled out and will apply to all newly written data.
Snowflake is also creating a new dashboard, which will allow customers to better monitor and understand their usage expenses. Now in public preview, the dashboard will also make it easier for organizations to manage their Snowflake accounts.
Look for more to come about the changes to Snowflake
As Snowflake’s features expand, so do the platform’s capabilities for streamlining your cloud data needs. Many of the advances announced at Snowflake Summit are still in progress, but the potential benefits remain clear.
Your data analytics processes stand to become more efficient, streamlined, and flexible. Plus, through Snowflake’s ongoing improvements, your cloud warehouse needs stand to grow more cost effective as well.
We’ll be paying attention as these developments continue to roll out and take shape. If you’re interested in doing more with your data warehouse, then your organization should be keeping watch on where Snowflake is headed too.