Within SAS solutions focused on improving performance and response time analytic environments highlights the effort with the proposed SAS In-Database. This proposal is accompanied by other high-performance solutions such as Grid Computing and SAS In-memory, which also improve performance, help manage the workload and improve the scalability of analytic environments.
The SAS In-Database solution reduces the need to move large volumes of data between SAS and DBMS solutions. Perform data processing, analytical or reporting processes within the database where this information resides reduce unnecessary data movement and provides the functions of data governance. Moreover, takes advantage of MPP architecture of database and data Warehouse environments to improve scalability and processes performancce.
In particular, emphasizes the integration of SAS and Teradata. This agreement has created SAS and Teradata Analytic Advantage Program, combining the analytical power of SAS with the benefits of Teradata data warehouse environments.
The solution reduces costs in the preparation of data for analytical processes and improves time in obtaining analytical results. Optimizes the creation of analytical models, converting SAS models in database objects. It reduces the data movement, redundancy and data refresh problems.
The program offers three levels, depending on the analytical needs of the organization:
Express: A first level to facilitate the use of statistics to solve business issues. Integrates SAS modules SAS Analytics Pro , SAS / ACCESS Interface to Teradata and SAS Enterprise Guide .
Advanced: It offers more in-database analytics solutions, like rapid creation and training of statistical models. Integrates SAS modules: SAS SAS Analytics Pro , SAS / ACCESS Interface to Teradata and SAS Enterprise Miner and SAS Scoring Accelerator for Teradata.
Enterprise: A complete solution that offers a complete portfolio of analytical models. Centralizes the creation, development, implantation and management of models. Automates analytical processes to support making decissions at all levels of the organization. Integrates SAS modules: SAS SAS Analytics Pro , SAS / ACCESS Interface to Teradata and SAS Enterprise Miner , SAS Model Manager SAS Scoring Accelerator for Teradata.
As a simple example of integration of the Teradata-SAS procedures, SAS documentation, offers the following case:
A method of use is fairly widespread is SAS PROC FREQ. This procedure performs a simple descriptive statistics based on a SAS data set.
Let's look at a case:
proc freq data = customer.credit_data;
table state * credit_score;
The specification DATA= indicates the procedure FREQ PRO that the input table is custormer.credit_data where customer indicates the database schema. credit_data is the name of the table in Teradata DB. The procedure obtains statistics from credit_score column values grouped by the state field.
To obtain the results, the unique values of the crédito_score column of the table crédito_data should be counted by the state field. PROC FREQ also obtained the minimum and maximum values.
For these in-database results, dynamically generates SAS SQL equivalent to procedure PROC FREQ. This SQL is sent to the module SAS / ACCESS, which in turn sends it to Teradata Database engine
select count (*) as f1, state, min (state) as f2, credit_score,
min (credit_score) as f3, max (credit_score) as f4
from customer.credit_score group by state, credit_score
