Business Analyst Direct Access to Data in Hadoop
Most companies know how to collect, store, and analyze their operational structured data. But new multi-structured data types are often too variable and dynamic to be cost-effectively captured in a traditional database schema using only SQL for analytics. Data scientists, business analysts, enterprise architects, developers, and IT managers are searching for new analytic solutions that can transform these huge volumes of complex, high-velocity data into pure business insight with new data-driven applications and analytics that give them a competitive edge.
Business analysts, in particular, need a new way to unlock and analyze massive amounts of multi-structured data that has been stored in ApacheTM HadoopTM without requiring complex Hadoop MapReduce programming and distributed processing skills. They need tools that allow them to use a standard business query language like SQL to invoke powerful MapReduce functions for analyzing data that has been captured and/or refined in the Hadoop environment. They also need access with existing BI and reporting tools, lowering overall cost and time to value.
Aster SQL-H empowers business analysts to directly access vast amounts of data in Hadoop for advanced analytics. With SQL-H, analysts can use common BI and reporting tools which leverage their business knowledge and SQL skills. They can access data in Hadoop directly, easily join it with data as needed in Aster Database, and leverage the analytical power of SQL-MapReduce® and business-ready analytic functions and applications such as click-stream analysis, marketing attribution, and graph analysis.
Low Latency, Standard and Easy to Use
Data stored in the Hadoop Distributed File System (HDFS) is processed in batch mode using Hadoop MapReduce jobs. These Hadoop MapReduce jobs are usually complex to develop and require the expertise of developers and data scientists that have a deep understanding of procedural programming.
Aster SQL-H creates a higher-level of abstraction by allowing ANSI standard SQL queries against Hadoop data. It leverages the power and flexibility of Aster SQL and SQL-MapReduce to provide business analysts with a low latency, interactive data discovery environment through their existing BI tools and SQL-MapReduce functions.
Supercharge Analytics from Hadoop Data
In addition to extending Aster’s SQL and SQL-MapReduce processing to Hadoop data, business analysts can use the suite of pre-packaged functions in the Aster MapReduce Analytics Portfolio to quickly build analytical applications on click-stream analysis, marketing attribution, fraud detection, time series, market basket, graph, and advanced statistical analysis on data stored in Hadoop. SQL-H radically shifts the focus of the business analyst from understanding Hadoop MapReduce to delivering business value.
Best Hadoop Integration with Existing Systems
Aster SQL-H easily integrates enterprise data from databases and/or data warehouses with Hadoop data for better SQL analysis. When doing data discovery analytics like clickstream or customer pattern analysis, business analysts can run SQL queries that join weblog data stored in Hadoop with customer data from an enterprise data warehouse.
Integration Based on Hadoop Standard
SQL-H leverages Hadoop standards to provide a business user abstraction layer to HDFS. SQL-H interfaces with the Apache HCatalog project to provide a mechanism for users to directly access the data in Hadoop from Aster. This enables Aster-managed communication with Hadoop nodes, to intelligently read just the data needed from Hadoop for SQL queries and SQL-MapReduce functions in Aster. Unlike competing methods using external tables, SQL-H leverages the metadata library of H-Catalog, easing administration.
To learn more about Aster SQL-H, contact us by phone at 1.888.Aster.Data or by e-mail to firstname.lastname@example.org.
Harnessing the Value of Big Data Analytics
The Next Generation of Big Data Analytics
Tasso Argyros, Co-President, Teradata Aster, discusses a seamless way to bridge the Hadoop and SQL gap