SAS® High-Performance Optimization
While data and models continue to grow in scope, detail and complexity, decision makers need answers faster than ever. SAS High-Performance Optimization enables you to model and solve optimization problems that are very large or whose other characteristics make them difficult or cumbersome to solve. You can develop and solve sophisticated, challenging models more efficiently and effectively to produce better, more accurate, and timelier decision guidance.
SAS High-Performance Optimization is available for execution in a highly scalable, distributed processing architecture.
With the ability to solve these kinds of optimization models quickly, you can perform more frequent modeling iterations and use sophisticated analytics to get answers to questions you never thought of or had time to ask. These features are available on Pivotal (previously Greenplum) and Teradata appliances, as well as on commodity hardware.
- Quickly and confidently seize new opportunities, assess alternatives thoroughly and efficiently, and make the right choices.
- Use advanced modeling and solution techniques, and perform more model iterations to get better answers to your difficult questions.
- Generate insights at breakthrough speeds for high-value and time-sensitive decision making.
- Take advantage of a highly scalable and reliable analytics infrastructure to test more ideas and investigate more scenarios.
High-performance local search optimization
- Optimizes a user-defined objective subject to linear and nonlinear constraints.
- Allows continuous and integer variables.
- Uses genetic algorithms, as well as other global and local search techniques, in parallel to solve problems.
- Provides a decomposition algorithm that works well for certain classes of large, structured linear and mixed-integer optimization problems. The algorithm decomposes the overall problem into a set of component problems that can be solved quickly.
- Provides a multistart capability that increases the likelihood of identifying a globally optimal solution. It can be used to select and begin optimization from each point. The best solution found among all starting points is reported.
- Parallelized option tuner functionality in PROC OPTMILP helps identify the best option settings for submitted problems. The tuner searches among possible option combinations, runs multiple optimizations in parallel with different settings, and finds the best set of option values.
SAS® Client Environment
The "high-performance client" must have the following SAS software separately licensed in order to use the high-performance component associated with the high-performance client:
- Base SAS/SAS 9.4 client running on one of the following platforms:
- HP/UX on Itanium (64-bit): 11iv3 (11.31)
- IBM AIX (64-bit) on POWER architectures 7.1
- Linux x64 (64-bit): SuSE 11 SP1; RHEL 6.1; OEL 6.1
- Solaris on x64 (64-bit): Version 10 Update 9; Version 11
- Solaris on SPARC (64-bit: Version 10 Update 8
- Windows x64 (64-bit): Windows 7 for x64 SP1 (Enterprise, Ultimate and Professional); Windows 8 for x64 (Enterprise, Standard and Pro)
- License for one or more of the following running on one of the platforms listed above:
- SAS ® High-Performance Statistics
- SAS ® High-Performance Data Mining (requires SAS ® High-Performance Statistics)
- SAS ® High-Performance Optimization
- SAS ® High-Performance Econometrics
- SAS ® High-Performance Forecasting (requires SAS ® High-Performance Econometrics)
- SAS ® High-Performance Text Mining
- Relevant SAS/ACCESS engine(s)
SAS® High-Performance Compute Environment
- Linux x64 (64-bit): Novell SuSE 10+; RHEL 5.5+; Oracle Linux 5.5+
- Teradata 720 compute appliance
- Pivotal (previously Greenplum) compute appliance
- Oracle Exalogic/Oracle Big Data Appliance/Oracle Commodity Hardware
- HP, Dell or IBM Commodity Hardware running Apache Hadoop 0.23 release
- HP, Dell or IBM Commodity Hardware running Cloudera CDH 4.2
NOTE: SAS High-Performance Forecasting runs ONLY on Pivotal Greenplum Database and Teradata.
Please consult your local SAS sales representative if you have questions about your platform requirements. Also, for more detailed information, please visit our support site at http://support.sas.com/resources/sysreq/.
Ready to learn more?
Call us at 1-800-727-0025 (US and Canada) or request more information.