Loading…
  • Sessions appear in the color of their primary track
  • Sessions can be filtered using Products on the right
  • Use the Search bar for more flexibility
See this link for hints on how to search the schedule or sign up for sessions
Back To Schedule
Tuesday, March 20 • 2:30pm - 3:20pm
Accelerating Joins and Aggregations on the Oracle In-Memory Database
Tweet Share
Feedback form is now closed.
Analytic workloads are dominated by joins, aggregations, and scans. In-Memory columnar databases have optimized scans using compressed data formats and SIMD vectorization techniques, but have made little impact to the rest of the query execution plan. The Oracle Database In-Memory (DBIM) Option introduced new SQL execution operators that accelerate analytic queries by optimizing aggregation over joins for star schemas. Group-by expressions are pushed down into the scans of dimension tables, creating a unique key per distinct group called a Dense Grouping Key (DGK). A Key Vector is allocated that maps join keys to DGKs, which is used to filter non-matching rows during the fact table scan. Passing rows are then aggregated directly on compressed codes into DGK-indexed result buffers using SIMD. Our solution replaces traditional join and group-by processing (bloom filters, hash table build/probe, etc) with blazing fast inlined scan operators, improving query elapsed time by as much as 10X.
Speakers
avatar for Shasank Chavan

Shasank Chavan

Vice President, In-Memory Technologies, Oracle
Lead an amazing team of brilliant engineers in the Database organization responsible for developing Oracle's In-Memory Column Store. We currently are hyper-focused on leveraging emerging hardware technologies to build the next-generation data storage engine that powers the cloud... Read More →
avatar for Ekrem Soylemez

Ekrem Soylemez

Architect, Oracle
I have been working in Oracle RDBMS development for 23 years. During that time, I have worked on data storage, Oracle OLAP, and most recently the SQL engine.  My current focus is improving performance of analytic queries, particularly those that are run against In-Memory Columnar... Read More →

Tuesday March 20, 2018 2:30pm - 3:20pm PDT
2-Rm 102
  Data Warehousing and SQL