October 13, 2020
Modern Data Warehousing practices are evolving, and Snowflake is rising to meet the growing needs within the warehousing space. Snowflake is different from traditional data warehousing tools for several reasons which we will cover in this article, but primarily because it separates compute, storage, and management functions. Ultimately, this is important because it saves consumers money and considerably improves performance. To understand how Snowflake can save you money and improve performance, we’re going to tell you:
- what Snowflake is,
- what solutions it provides,
- who might be interested in Snowflake,
- where it fits in an existing enterprise environment,
- and how the architecture is different and why that matters.
Snowflake is a Cloud Data Platform that is delivered as a service with the goal of enabling any user to work with any data without limits on scale, performance or flexibility. It features compute, storage, and cloud services layers that are logically integrated but scale independently from one another, making it an ideal platform for many workloads. (1)
Over time, many technologies have attempted to solve the common problems associated with data management. The on-premises enterprise data warehouse (EDW) was the initial, popular option that came to be, and has delivered in performance for well defined, structured data. EDWs are easy to use and were rapidly adopted. However, these technologies fall short in supporting different data types and a multitude of concurrent workloads. First generation cloud data warehouses stepped up to offer better cost performance than traditional EDWs, however still retained the same issues with data types and concurrent workloads. On the other hand, big data solutions came into the mix and with them came the ability to handle structured and semi-structured data. While this was a win for the data management community, they had limited performance capabilities and were complex to use. All of these solutions require separate hardware for dev, test, and prod environments.
Snowflakes cloud data platform takes the best features of each of these predecessors. It has provided a solution that allows for scale, performance, and concurrency, all while alleviating the need for separate hardware for different environments. With these enhancements over traditional technologies, the industry has seen improvements in performance and speed, support for different types of data, and concurrency capabilities. Additionally, Snowflake offers security and data sharing features for their clients.
Based on customer testimonials and capability analysis, we believe that Snowflake could benefit most businesses. Specifically, the advantages we discussed could be well suited for:
- Companies interested in creating a cloud-first data strategy,
- Small to medium companies with limited data warehousing experience but are looking to get started,
- Large companies looking to maximize their efficiency and save on costly modernization efforts moving forward, and
- Any company looking to associate business usage with Total Cost of Ownership (TCO)
Snowflake sits where you’d expect a data platform to fit – right between your data sources and your front end analytics tools. There are scripts that can be run to load data directly into the platform, but Snowflake has several partners that work really well with the technology to simplify this process to handle orchestration of data loading. From an analytics perspective,most major analytics platforms/tools (including Power BI, Qlik Sense, and Tableau) have built-in connectors for using Snowflake as a data source.
Personnel responsibilities within analytics are adjusted with Snowflake in use. For example, administration will still fall to a technical resource; someone who understands data architecture. This person or team would likely be responsible for making sure data is loaded into the platform, proper security is applied, and data made available to business users. The need for much of the traditional database administration tasks goes away and this type of role might fall to a Data Engineer or ETL Developer.
There will still be a need for data modelers that can take that raw data and transform it into business data that can be consumed by business end users. For reporting, savvy business users may interact with the interface and could have access to write queries. For most, users would primarily be expected to interact with the platform through the BI layer that is established within the organization (i.e., Power BI, Qlik, Tableau, etc.).
With over 3,400 customers since the company’s genesis in 2012, Snowflake has experienced massive growth and acquired many household names as active clients in addition to distinguishing itself as one of Gartners’ Data Management Solution Leaders. With growing demands in the data warehousing space, we anticipate significant growth moving forward.
Want to learn more about how Snowflake solutions can help your business? Book a Modern Data Management Discovery Assessment and we will provide you with a custom readiness assessment that outlines our top recommendations for implementing a modern data management approach.
Written By: Ellari Hillard