How to Use a Semantic Layer to Deliver Actionable Insights at Scale
0 likes696 views
The document discusses the implementation of a semantic layer to enhance enterprise business intelligence (BI) and artificial intelligence (AI) by providing agility, consistency, and control. It highlights the value of a universal semantic layer in optimizing cloud analytics, serving as a single source of truth for enterprise metrics, and modernizing legacy OLAP systems. Performance improvements are quantified through benchmarks, showing significant enhancements in query execution times and reduced SQL complexity with the use of AtScale.
How to Use a Semantic Layer to Deliver Actionable Insights at Scale
- 1. How to Use a Semantic Layer to Deliver Actionable Insights At Scale Unlocking the value of your data with a Semantic Layer Dataversity Dave Mariani, CTO, Founder July 26, 2022
- 2. 2
- 3. A semantic layer provides the agility, consistency, and control needed to scale enterprise BI and AI 3 SL SL SL SL DI SL DI SL DI SL DI BEFORE AFTER
- 4. Where does a semantic layer fit in the stack? SEMANTIC LAYER DATA CONSUMPTION DATA MODELING DATA API LAYER DATA PREPARATION RAW DATA BI TOOLS AI/ML TOOLS APPLICATIONS CLOUD DATA WAREHOUSE DATA LAKE ENGINES SAAS APIs DATA TRANSFORMATION TOOLS DATA LAKES SAAS APPLICATIONS DATA CATALOG 4
- 5. How to Use a Semantic Layer on Big Data to Drive Al & BI Impact © 2022 AtScale. All rights reserved. How to Use a Semantic Layer on Big Data to Drive Al & BI Impact © 2022 AtScale. All rights reserved. 5 The Semantic Model ➔ Business Friendly ➔ Single Source of Truth ➔ Consistent Governance ➔ Performance Optimized
- 6. The Value of a Universal Semantic Layer 6
- 7. How is a semantic layer used? CLOUD ANALYTICS OPTIMIZATION Speed of thought analytics on live cloud data, agile data integration with minimal data movement. ENTERPRISE METRICS STORE Single source of governed enterprise metrics that can be self-served from any AI/BI tool. BRIDGING AI AND BI Common view of data assets and platform for publishing AI-generated insights to decision makers. OLAP MODERNIZATION Migrate legacy OLAP (e.g. SSAS) models to modern cloud-first infrastructure. 7
- 9. TPC-DS 10TB Benchmark: Improvements with AtScale 9 1. Elapsed time for executing 1 query five times 2. Elapsed time executing 1 (x5), 5, 25, 50 queries 3. Compute costs for cluster time (Redshift, Snowflake) or bytes read (BigQuery) for user concurrency test 4. Complexity score for SQL queries for number of: functions, operations, tables, objects & subqueries (AtScale = 258, TPC-DS = 1,057) Test Improvement Factor with AtScale BigQuery Redshift Snowflake Synapse Databricks Query Performance1 4x Faster 11x Faster 3x Faster 6x Faster 4x Faster User Concurrency2 11x Faster 31x Faster 12x Faster 11x Faster 15x faster Improved ROI3 3x Better 4x Better 16x Better 2x Better 6x Better Complexity4 76% less complex SQL queries
- 10. www.atscale.com