Starburst + VaradaStarburst Acquires Varada: Setting a New Standard for Data Lake Analytics
Your data lake becomes instantly operational & interactive
Varada’s dynamic and adaptive indexing technology ensures queries will run at interactive response time. The more selective queries are, the greater impact indexing will deliver. Varada automatically analyzes and detects which datasets to accelerate, and then applies the optimal index. Data architects maintain full control in prioritizing workloads, defining budgets and performance requirements.
Varada enables queries to run on any column at its source granularity, unlocking the true value of data. To ensure performance, many enterprises compromise on accessing all their available data and settle for isolated data silos that have been prepared and modeled to enable speedy queries. Varada instantaneously transforms your available data into operational data so you can say goodbye to lengthy data preparation projects and leverage x10 more data.
Varada’s machine-learning optimization tools continuously track cluster performance and data usage. Varada monitors queries on a workload-basis to see which tables are used the most, how queries are running and where bottlenecks form, and automatically adapts the operational dataset. Varada also leverages advanced cost-based optimization and elastic scaling to ensure the best possible resource utilization.
Varada is deployed as a private managed service or as a Presto / Trino connector, running in your Virtual Private Cloud (VPC), so that data is not duplicated and does not need to leave your account. Varada can seamlessly access all your existing data sources and serves any SQL data consumers out-of-the-box.
Deliver dramatic query acceleration and concurrency improvements on your existing Trino clusters, with Varada’s autonomous indexing and smart cache management.
Everything else remains as-is…
The community edition is free (up to 4 nodes)!
Accelerate query performance and support high concurrency directly on your data lake, with zero efforts and zero data modeling.
Read more
Deliver consistently interactive dashboard performance, drill downs, and high concurrency, running on hundreds of granular TB stored in the data lake.
Read more
