At AtScale, we’re committed to helping enterprises realize the value of their legacy platform investments and capitalize on the speed at which data is proliferating. Data architecture modernization is a major business driver for our customers and it's our primary focus to enable them to modernize without disrupting critical business functions. AtScale’s newest release 2019.1 helps us stay true to that mission, specifically addressing the challenges associated with migrating from on-prem to cloud, RDBMS to Hadoop, and legacy BI to advanced analytics.
Better BI user support to use data in the context of business relevance
New support for PowerBI MDX Connector
The PowerBI trend continues to dramatically expand into enterprises. In particular, business analysts and citizen data scientists are attracted to Microsoft’s latest BI tool for its ease of use and robust BI capabilities—especially for the support for dimensional awareness when connecting to your data source.
But what challenges does this present? Access to rows and columns is a good start for ETL developers and data scientists. But tables and columns lack the business context which is inherent in AtScale virtual cubes. The business user and the citizen data scientist want to use data in the context of business relevance. Hierarchies, logical drill-down, grouping elements—this is the difference between 200 columns versus the grouping related query elements which drill down from Product Super-Category, Product Type and Product Line.
PowerBI now connects to AtScale and displays the AtScale cube with all the logic that makes the cube easy to understand and use. No more need to sift through hundreds of columns, you now have dimensions and key measures to work with.
Deeper Data Warehouse integration that dramatically increases the ease and speed for support
Data Warehouse SDK
When AtScale supports a data warehouse technology, we strive to do the deepest possible integration, supporting native security and performance features versus simply being able to generate the correct SQL dialect. In some cases generating the proper query and using the right underlying tech can result in a massive reduction in resources used, and in a cloud data warehouse, this could mean making a query much less expensive. An example of this in Teradata is single AMP vs all AMP queries utilizing tech such as sparse maps.
Because these implementations are much more sophisticated, AtScale introduces a formal Data Warehouse SDK which is an abstraction that formalizes functional areas where developers can take advantage of the infrastructure provided to build faster and less expensive support for new data warehouses. This SDK and approach to integration benefit single data warehouse implementations as well as federated, virtualized implementations.
The Data warehouse SDK means AtScale will support Data Architecture Modernization by matching the pace of innovation in the cloud space.
Precision and speed at scale for decimal
As the big data and cloud industries raced towards adoption, support for certain data-types lagged behind. You would expect this for brand new data-types, but in some cases, it was the opposite. Decimal support across AtScale and underlying platforms are now fully supported. The precision that many operational analytics require can be stored and calculated without the fear of losing precision for the sake of data scale, query speed or storage cost.
The March towards Zero Trust
SAML 2.0 Support
Security and governance are becoming a much bigger issue for all BI vendors. Meeting regulatory compliance will move from being "really should have this" to "must have this" over the next 2-5 years, especially as the notion of a perimeter dissolves and data is shared outside the enterprise.
In 2019.1, AtScale takes a major step towards full Zero Trust Architecture support with the addition of SAML 2.0 support for Okta, Auth0, Shibboleth, Ipsilon, OneLogin, and Google. SAML is a key cloud-native federated identity technology supported by all major cloud vendors, and essential for hybrid cloud/on-premise use cases.
Support for multiple SAML for multiple providers and an easy way to upload Identity Provider Metadata for federated identity support.
The key benefits to Zero Trust Architectures include:
- SSO via OAuth/SAML — Never share passwords, never require the user to give credentials to your applications.
- Multifactor — Make users authenticate and give them a challenge on a separate channel. We do this even for popular tools such as Excel that do not support it natively.
- Integration with External Directories — via LDAP/Active Directory or Cloud provided Identity Management solutions.
- Role Based Access Control — We do great things here, best in the industry.
- Validation for Every Device — Confirm new devices accessing data to limit the surface area.
We predict data privacy, security, and governance will continue to be the main headline for enterprises through the rest of 2019 and through 2020. Many banks require their vendors to become compliance certified partners, which requires reporting on security-related metadata as it relates to access to data. Like the old saying “As California goes, so goes the nation”, we like to think “As Wall Street goes, so goes the industry” and we will see this trend for more stringent security requirements move forward quickly.
When enterprises move towards a single, business-driven source of truth for BI and ML, the introduction of better self-service solutions will drive more and more consumption of data.
The more you enable users to be successful with data, the more data they will want, and increasing the availability of data always has to be matched with improved comprehensive security.
We’re finding that Zero Trust worthy use cases exist inside and outside of finance. Many retail customers are turning on to the incredible value of their data for their suppliers, which means new inter-enterprise use cases are becoming both top line revenue generators as well as hints for optimizing product offerings and marketing campaigns.
Improved Usability and Agility
We’ve always known it was not easy to use our Snapshots Panel when it was outside of the cube. It’s a much more natural workflow to see the exact details of your model and then take a snapshot. Using data from our opt-in secure customer telemetry, we checked the number of users that were taking Snapshots. We realized that even though taking snapshots was not easy, many of our customers had taken hundreds. This realization led us to improve the functionality for daily use.
Our analysis showed Snapshots are a daily part of many users workflow, so we made this feature more convenient and easier to use.
Enhanced User Specific Landing Page for Published Cubes
For the landing page, we wanted to provide something relevant for BI users just after login. Along with the new project and cube information pages that surface more details on fact tables, dimensions, measures, and partitions, according to our user research, BI users can benefit from improved visibility into related published activity and connection information up front.
Quick Access to BI Tool connection strings and other Project metadata was the goal of redesigning the Project Landing Page.
Ultimately Achieving Business Value with Confidence
In this era of unprecedented data transformation, virtualizing your data warehouse makes it work for today, tomorrow and the future, all while capitalizing on the analytics revolution that requires a new approach to security, governance, and agility. Our product roadmap enables the enterprise to safely realize that investment with our unified view and virtualized approach to data migration for spend, performance, and efficiency gains. AtScale 2019.1 introduction proves our commitment with its data platform SDK for accelerated onboarding of new data platforms. The platform update introduces native multi-dimensional support for Microsoft Power BI, data warehouse platform support for Teradata and PostgreSQL, a software development kit (SDK) for accelerated on-boarding of enterprise data platforms including Oracle and advanced query and security capabilities. We help you transform your data warehouse for the future, eliminating the data scaffolding that currently confines and constrains it. 2019.1 is the first step this year. Stay tuned for our quarterly updates and enhancements!