ParAccel

Time to Create an Unfair Business Advantage with Data Warehousing

High performance for data warehousing, analytics, and business intelligence has never been so easy. The ParAccel Analytic Database™, with its innovative massively parallel processing (MPP) and columnar design, is a real breakthrough for companies that want to extend their analytic performance advantage by gaining faster access to data and the ability to accomplish in seconds what used to take hours.

High Performance Data Warehousing and Analytics: Why ParAccel's MPP, Columnar Database is Most Efficient

  • Only relevant columns are retrieved (A row-wise database would pull all columns and typically discard 80-95% of them)
  • All data warehouse operations are done in parallel (a non-parallel DBMS must scan all of the data sequentially)
  • Adaptive compression makes disks faster and reduces decompress effect
  • A memory-centric design maximizes opportunity for in-memory processing
  • Query compilation reduces the instruction set and makes CPUs more efficient
  • A custom protocol optimizes inter-nodal communications

Row-based vs Columnar Database Processing

Data warehousing, analytics and business intelligence require heavy read activity within a database management system. The following illustration highlights ParAccel's columnar scan efficiency advantage over a row-based DBMS — in this case, 50X. Add ParAccel's massively parallel processing and compression for even greater speed.

Data Warehouse Query of U.S. Census Data
"What is the average age of people by state?"
Common Row-based Database ParAccel's Column-based Database
Number of Records 300M people@ 1k bytes/person 300M people@ 1k bytes/person
Scan Volume 300GB (all columns) 6GB (2 columns)
Efficiency 2% 100%
Performance Minutes Seconds