Analytics

Typical Issues:

  • High latency for analytics on terabyte to petabyte data sizes
  • Business analysts are forced to sample data risking undiscovered patterns, limited insight and missed events
  • Advanced analytic techniques such as pattern, time-series, clustering, graph, and market basket analysis are not available to analysts in a timely manner or at all

Causes:

  • Parallel programming is too complex for most organizations
    • Developers must have specialized skills to program for parallelism
    • Native Hadoop MapReduce is not business-analyst friendly
  • Advanced analytic toolsets rely on a high latency data pipeline
    • Analysts must prepare and move data out of the data warehouse for advanced analytics
  • Advanced analytics stress the limitations of standard SQL alone
    • Multiple self-joins slow query performance
    • Minor application/analytics modifications require significant SQL code changes

The Solution:

  • SQL-MapReduce®
    • Integrates SQL with MapReduce parallel processing framework
    • Power of MapReduce with familiarity of SQL
    • Automatically parallelizes analytic applications without requiring design for parallelization
  • Embedded Analytics
    • 100% of analytic computations run in-database, close to the data, eliminating massive data movement
  • Aster Data Analytic Foundation
    • Suite of pre-built SQL-MapReduce functions that provide high performance analysis for advanced analytics including pattern,time-series, clustering, graph, and market basket analysis
    • Pre-defined functions are modifiable by the business analyst simply through parameter changes - no additional time-intensive coding required
  • Hybrid Row/Column Database
    • Unified SQL-MapReduce computation layer can access data optimized for performance through either a row store, column store, or a combination of both
    • Both row and column stores are first class citizens in the database architecture, sharing a unified set of data services for ease of management and seamless data access
    • Data storage recommendations provided based on workload to ease modeling effort
Top Picks
Whitepaper: A New Approach for Large-Scale Data Management and Data Analysis
Whitepaper: Deriving Deep Insights from Large Datasets
Forrester Report: In-Database Analytics: The Heart of the Predictive Enterprise
Now that all our data is in one place, we can understand customer interactions across our entire [retail/ online/e-reader] ecosystem. (MapReduce helps researchers) see trends more quickly than possible in systems only using massively parallel processing.

Barnes&Noble
Marc Parrish, Vice President,
Direct Marketing
With Aster nCluster ... our data load time has decreased by over 95 percent, and our most important queries complete in seconds or less.

ShareThis
Tim Schigel, CTO