The leadership team of a Fortune 50 retailer needed to connect thousands of internal and external analytics consumers and move their data from an on-premise enterprise data warehouse to Google BigQuery without disrupting their business. With AtScale’s semantic layer platform, they were able to seamlessly move analytics infrastructure to the cloud, achieving the scalability and efficiency to support both growth and analytics agility.
reduction in query costs
migrated in weeks
queries per day
Scaling up analytics and modernizing data infrastructure.
A Fortune 50 retailer launched an initiative to modernize their analytics infrastructure with the primary goal of increasing the flow of data-driven insights that could lead to improved margins, optimization of product mix, and better inventory management.
Their challenge was to enable better analytics at scale while ensuring efficiency and consistency across a broad audience of data consumers. With thousands of users performing analytics using a diverse set of legacy platforms, including SQL Server Analysis Services (SSAS), Teradata, and Hadoop, the existing infrastructure was expensive and could not scale at the rate of their business. To empower their users, the data team needed a scalable semantic layer solution that could serve the needs of internal users as well as suppliers that rely on a shared view of inventory. The solution needed to scale, needed to support security and access control policies, and needed to support the organization’s migration from on-premise legacy data platforms to a cloud data warehouse.
Applying a semantic layer to accelerate analytics performance and improve cost-efficiency
Initially, the business partnered with AtScale to replace traditional SQL Server Analysis Services (SSAS) OLAP instances. The AtScale semantic layer delivered the analytics performance of SSAS without the complex data engineering and the need to extract and transform data to maintain traditional OLAP “cubes.” This initial implementation was done with on-premise Hadoop data.
As the organization transitioned to Google BigQuery, they were able to leverage AtScale’s virtualization-based approach to seamlessly transition analytics with no interruption to the business. Within a single weekend, the data team was able to redirect existing AtScale models to the new cloud data repository, enabling existing reports, dashboards, and applications that were based on AtScale to continue operating with no changes.
Better data access with millions of dollars saved
In the first stage, this organization leveraged AtScale to transition SSAS-based analyses from Hadoop to Google BigQuery. Internal business users and external partners were able to run the exact same analytics as they had before the migration without realizing that the data store they were querying against had shifted to the cloud. They were able to continue their modernization by leveraging AtScale to manage cloud costs, ensure predictability of performance, and create the scalability needed to accommodate growth in the years to come.
By working directly in Google BigQuery, AtScale reduced the cost of a query by 91%, empowering the organization to re-allocate funds to increase the number of value-creating analysts to better support the needs of the business. Additionally, they were able leverage efficiencies to increase data retention from 3 months to 3 years (a 1200% increase). Query times were reduced substantially while radically increasing efficiency enabling the organization to support the 17,000+ queries per day being initiated by their large constituency of internal and external users.
Self Service BI, Analytics Modernization, Cloud Migration
Manage cloud query costs, Support user growth, Increase data retention
Key product components
Google BigQuery, Excel, Tableau