How We Make BI Work on Big Data

A Modern Architecture for Business Intelligence

Atscale is designed to support key requirements of modern big data analytics.

We support hybrid data access to the most common big data platforms and provide open data interfaces (SQL, MDX, or API) to virtually any analytics front-end.  AtScale’s self-tuning engine ensures interactive performance for data consumers on even the largest data sets.

Designed for universal semantic model management, AtScale provides an easy-to-use interface to model virtual cubes, while system metadata maintains a graph of data lineage all the way to the underlying raw data. With native support for complex business analytics, including distinct counts, percentiles, parallel periods, multi-fact models, and complex MDX calculations, AtScale can support your most complex analytic and OLAP use cases.

"AtScale was born to deliver on the promise of making big data useful for the business."

- Jen Underwood, Impact Analytix

Business intelligence tools and drivers

Standard connection to your favorite BI tools
AtScale enables current business processes by allowing your analysts to use the BI tools they already know and love, but at sub-second speeds on live big data.

  • Use common BI tools like Excel, Tableau, Qlik, or Microstrategy
  • Connect with standard BI and big data drivers you already use
  • Choose to use API, JDBC/ODBC, OLE DB, or XMLA

Learn more in the AtScale datasheet

Universal semantic layer

Common metadata repository is a BI force multiplier
With a universal semantic layer and robust metadata and lineage tracking, AtScale enables business logic to exist directly on top of the big data cluster and of BI tool-specific semantics.  As a result, the AtScale semantic layer can be used by your favorite BI or analytics tool.

  • Share business definitions and logic across your BI tools
  • Ensure shared security policies for all users
  • Track data lineage and audit usage across all data assets

Read our universal semantic layer paper

"The Switzerland of semantic layers."

Krishna Roy, 451 Research

Dimensional calculation engine

Multi-dimensional queries on big data made simple
While big data platforms excel at low-cost distributed storage and parallel data processing, interactive support for BI-style queries remains a challenge. Additionally, multi-dimensional analytical queries often demand complex OLAP-style calculations and functions. AtScale’s dimensional calculation engine enables these complex business queries directly on your big data.

  • Multi-pass query support
  • Semi-additive metrics like first & last child
  • Multi-level metrics, such as plan vs. actual
  • Atomic drill-through to record level data

Read more on multi-level metrics

Machine learning performance engine

Automated query optimization
With AtScale, your queries run fast on big data because we automatically generate and maintain your aggregates. Advanced machine-learning algorithms optimize your BI query workloads on-demand to deliver the performance you and your users have come to expect from BI, multidimensional and OLAP systems.

  • Automatic and dynamic aggregate creation
  • Option for manual aggregate override
  • Easy aggregate maintenance
  • Enabled for both on-prem and cloud big data platforms

Data abstraction layer

Abstracting dimensionality into big data
With AtScale’s patented approach to data abstraction you easily create complex, multidimensional analytical models directly on top of a variety of Big Data types (flat files, relational data structures, or semi-structured). Now Enterprises can model the most complex business structures for analysis – live on the data in your on-prem and cloud big data platforms.

  • Dimensions with multiple hierarchies
  • Support for multi-fact models
  • Complex calculated metrics
  • Pre, post, and mid-query aggregation functions
  • Works with big data (on-prem and cloud) Hadoop, BigQuery, etc…

Learn how we handle multi-Fact models

"AtScale’s no-ETL and no-data movement approach is simply a gamechanger. This application should be required for anyone who wants to do BI on Hadoop."

Kevin Johnson CEO, Ebates

Security

Enterprise-level access control, monitoring, and management
AtScale works with enterprise big data deployments to offer data governance and security for the data residing in Hadoop and other big data Platforms. Administrators have complete control over who can access which data across big data in the Enterprise.

  • Role-Based access control
  • Pluggable SASL authentication
  • User and query audit trails
  • Delegated authorization support

Learn more about AtScale security

Visual Modeler

Easy, intuitive BI model and cube design
Business users get to interact directly with data in Hadoop and other big data systems via the AtScale Design Center web app. Data analysts can design virtual analytical (OLAP) cubes using familiar workflows and intuitive drag-and-drop interactions.

  • Intuitive modeling concepts
  • Rich multi-dimensional (OLAP) support
  • Collaborative cube design and model lifecycle management
  • Fast, scalable distinct counts
  • Virtual data abstraction layer

Check our resources to learn more

COMPREHENSIVE FUNCTIONALITY

AtScale delivers an advanced set of features you won’t find together in one solution, anywhere else.

Drag and Drop Modeling

Built for the business analyst, drag and drop enables rapid cube design & deployment. Build sophisticated models without writing any code.

Multi-Tenancy Support

Supports multiple clusters, multiple environments, and multi-tenant data; significantly lowering IT maintenance costs.

Multi-User Environment

Cube development is a team sport - AtScale allows for realtime cube collaboration, with snapshot and restore functionality to easily track changes.

Security Integration

Integrates with LDAP and Active Directory, supports kerberos, and enables delegated authorization for data security that leverages existing infrastructure.

Works with SQL and MDX

AtScale is the only BI platform that supports both SQL and MDX queries against the same multidimensional model.

Hierarchy Support

Hierarchical analysis is core to any BI application, and AtScale delivers rich support including multiple hierarchies, role-playing, and sharing.

Multi-fact Models

Analyzing metrics across multiple facts is core to most advanced analysis - AtScale supports even the most complex multi-fact models.

Row Level Security

Make sure that the right users see the right data with support for dynamic user-to-attribute security policies.

Predictive Aggregates

Statistical models predict optimized queries even before a user executes it, while smart algorithms keep aggregates up-to-date with zero human intervention.

Many to Many Support

Many-to-Many models, a requirement for any advanced BI platform, are supported in to AtScale.

Incremental Updates

No need to rescan your entire data to update aggregates. Incremental updates look only at the most recent partitions of data.

Snapshots and Versioning

Built by developers for developers, we ensure that you can restore and dif any version of your model.