Our Platform

Varada on-demand SQL analytics tier allows any data practitioner to define on-demand any dataset of interest directly on the data lake.

At the heart of it, our inline indexing core will process across all dimensions and schemas to deliver breakthrough performance at a petabyte scale.

what We Do image

What We Do

Lightning-fast analytics on the data lake

Varada is a data-serving layer for data lakes that can materialize anything without building an ETL pipeline. Constantly synchronized with the data lake, our materialized views can serve sub-second response times on complex analytics.

Securely running in your account and VPC

The Varada solution is deployed within your secure cloud environment, allowing you to easily meet security and regulation requirements. Data and cluster access can be easily controlled with standard cloud services running in your account.

Breakthrough architecture

Varada’s unique materialized view, data layout and elastic, distributed cluster architecture are based on SSD and NVMf. Data and metadata nanoblocks are processed with zero latency and without bottlenecks across clusters created on-demand.

Standard SQL and data formats

Varada is based on Presto, the world’s leading virtual data warehouse that was designed from ground up for interactive analytics at any scale. Varada supports standard Presto SQL and JDBC/ODBC connectivity.

Built to productize data

Build data-driven products that maintain a predictable and consistent serving SLA on top of a data lake. With indexes prepared for any ad-hoc workload, an elastic cluster and enterprise-grade management, any new business demand can be met with minimal effort and time.

Fast analytics on any dimensionality

Indexes across all the dimensions in a view allow interactive dynamic segmentation, cohort analytics, funnel analytics and other complex ad-hoc queries on a data lake scale – on any set of dimensions, with any BI tool and without data modeling specific to the need.

Business agility

Varada guarantees business agility by eliminating the complexity of data preparation: instead of waiting months for data modeling and ETL development, have a fast analytic data-serving tier on demand.

Our Building Blocks

  • Cloud Data Lake

    Cloud data lakes provide a centralized data repository that is simple, cost effective, fault tolerant, accessible anywhere, and eliminates the need for data capture and ETL in data consumption operations like analytics and machine learning.

    We help you improve the way you uncover, use and operationalize data in data lakes by offering the only SQL analytics tier built to serve hot data directly from a data lake.

    Cloud Data Lake
  • Cloud Elasticity

    The ability to scale compute capacity on the cloud on-demand enables both business agility and significant cost savings. We leverage the cloud elasticity to enable you to handle any workload, any concurrency and any hot data volume in a cost effective manner.

    Cloud Elasticity
  • Apache Presto

    Apache Presto is a virtual data warehouse architected from ground up to provide the fastest SQL analytics on big data. Presto open source engine has been widely adopted and powers the analytics backbone of leading tech companies on hundreds of PBs.

    We chose Presto as our front end to provide the best SQL on the market, and we are a part of the Presto community.

    Apache Presto
  • NVMe SSD

    NVMe SSDs enable high bandwidth, small blocks, and extremely fast random I/O at almost zero latency with a new hardware and software stack designed for SSDs.

    Our unique materialized data layout is built to leverage the NVMe SSD zero latency data access patterns to provide unmatched performance and flexibility to any query.

    NVMe SSD