Varada has announced the availability of the Varada Data Platform, its flagship data virtualization product that features a dynamic indexing technology. This release comes on the heels of Varada’s Series A funding round of $12 million from September. Dynamic indexing “enables teams to balance performance and cost of queries at massive scale”, according to the company. Varada made the announcement at Solutions Review’s second annual BI Insight Jam.
Varada enables data architects to hasten and optimize workloads via dynamic analysis and adaptive indexing. This technology ensures that queries will run at interactive response time. The platform then automatically analyzes and detects which datasets to accelerate, and then applies the optimal index. Varada transforms data into operational data in real-time, and enables queries to run on any column at its source granularity. Included machine learning optimization tools continuously track cluster performance as well.
The Varada Data Platform sits atop a customer’s existing data lake. Users get visibility into workload performance and cluster utilization so they can define workload priorities and business requirements. Adaptive indexing breaks data across any column into nan blocks and automatically chooses the most effective index for each nano block based on the data content and structure. This makes queries fast without the need to model data or move it to optimized data platforms.
Varada currently runs on AWS and supports reserved, on-demand and spot instances. Pricing is per-node, based on a predefined scaling group. Enterprise support for the Varada Data Platform is also available from the provider. It supports an array of data sources and formats with support for Google Cloud and Microsoft Azure coming soon.
In a media statement about the release, Varada CEO Eran Vanounou said: “The beta period for this product has proven two things. First, that organizations are desperate for a way to simplify data ops management while getting the cost of query acceleration under control. Second, the path we’ve chosen is striking a chord: Varada is a ‘zero data ops’ approach that eliminates data silos by serving many workloads from one platform. And because all queries will run atop the data lake, there is a single source of truth that eliminates the need to move or model data.”
Learn more about the Varada Data Platform.
Latest posts by Timothy King (see all)
- The Three Best Data Engineering Books on Our Reading List - April 8, 2021
- The 8 Best Data Engineering Courses and Online Training for 2021 - April 8, 2021
- Trifacta Launches Industry First Data Engineering Cloud - April 8, 2021