IBM® Netezza® Analytics Release 2.0.1 IBM Netezza Analytics Release Notes Part Number 00J2008-03 Rev.
Note: Before using this information and the product that it supports, read the information in “Notices and Trademarks” on page 29. © Copyright IBM Corporation 2011, 2012. US Government Users Restricted Rights – Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.
Contents General IBM Netezza Analytics Topics ...................................................6 IMPORTANT – Read First Before Installing This Release.....................................................6 Database Compatibility......................................................................................................6 Compatibility With Prior Releases.....................................................................................6 Compatibility With Revolution R Enterprise for IBM Netezza.
Notices and Trademarks.........................................................................29 Notices.............................................................................................................................29 Trademarks......................................................................................................................31 Regulatory and Compliance............................................................................................
General IBM Netezza Analytics Topics General IBM Netezza Analytics Topics IMPORTANT – Read First Before Installing This Release Database Compatibility This release of the IBM Netezza Analytics (referred to as Netezza Analytics in the remainder of this document) supports Netezza systems that run release 6.0.5P5 or later. If your Netezza system is using an earlier release, you must upgrade it before using this release of Netezza Analytics. Compatibility With Prior Releases Release 2.
New Features in Netezza Analytics Release 2.0 Call Interface Changes to Analytic Functions The call interface to many analytic functions has changed in this release.
Call Interface Changes to Analytic Functions The table below lists functions with a changed call interface and provides examples of the changes through sample calls. Only the changed portion of the parameter list is shown; the ellipse (…) in the sample call assumes that other required parameters are provided and have not changed from an older release to this release. The table also notes whether backward compatibility is available: Prior Release (old way) (Sample partial call) Release 2.
Call Interface Changes to Analytic Functions Prior Release (old way) (Sample partial call) Release 2.0 (new way) (Sample partial call) Backward Compatible? /Notes call nza..CUMULATIVE( 'X=somecol, ...'); call nza..CUMULATIVE( 'incolumn=somecol, ...'); No call nza..DENSITY('X=somecol, ...'); call nza..DENSITY(' incolumn=somecol, ...'); No call nza..ENTROPY('X=WORKCLASS, ...'); call nza..ENTROPY(' incolumn=WORKCLASS, ...'); No call nza..JOINT_ENTROPY( 'X=age, Y=wage_per_hour, ...'); call nza..
Call Interface Changes to Analytic Functions Prior Release (old way) (Sample partial call) Release 2.0 (new way) (Sample partial call) Backward Compatible? /Notes call nza..SUMMARY1000( 'varlist=FIXED_ACIDITY; VOLATILE_ACIDITY; CITRIC_ACID; RESIDUALSUGAR, ...'); call nza..SUMMARY1000( 'incolumn=FIXED_ACIDITY; VOLATILE_ACIDITY; CITRIC_ACID; RESIDUALSUGAR, ...'); No call nza..T_LS_TEST ('X=petallength, Y=sepallength, ...'); call nza..T_LS_TEST( 'incolumn=petallength:X; sepallength:Y, ...
FPGROWTH Algorithms Renamed and Modified supported. The table below shows sample calls for both the older and new algorithm names. Prior Release (old way) Release 2.0 (new way) CALL nza..PREPARE_FPGROWTH( 'intable=nza..quant_sales, outtable=dset, tid=tid, item=idart'); CALL nza..PREPARE_ARULE( 'intable=nza..quant_sales, outtable=dset, tid=tid, item=idart'); CALL nza..FPGROWTH('intable=nza..retail, pfx=results, support=1 '); CALL nza..ARULE('intable=nza..
Modified Default Parameter Settings for DECTREE and REGTREE Parameter minsplit New Value 50 Old Value 2 Description The minimum number of instances in a node required for a split. If the number of instances in a node is less than minsplit, no further split is applied and the node becomes a leaf. Time Series Forecasting Support for Time Series is introduced in this release. A time series is a sequence of numerical data values, measured at successive, but not necessarily equidistant points in time.
Changes to the KMEANS Algorithm Clustering using Mahalanobis distance Normalized Euclidean distance Scoring with statistics of clusters and columns Automatic data normalization and standardization Enriched statistics See “KMEANS algorithm” and “Enriched Statistics for Clustering Models” in the IBM SPSS In-Database Analytics Developer's Guide for details of these new features. In this release there is a behavior change to the KMEANS algorithm.
Limited PMML Support for Analytic Models K-means Association rules (ARULE) Naïve Bayes Support is implemented with the following new analytic procedures: PMML_MODEL EXPORT_PMML This new feature is described in more details in the “PMML” section of the IBM SPSS InDatabase Analytics Developer's Guide. Logistic Regression and Generalized Linear Models (GLM) New to this release are algorithmic procedures to support GLM.
Changes to nzMatrix Simplified Matrix Multiplication The matrix multiplication procedure (GEMM) has been simplified. Previously, users chose whether to use GEMM or GEMM_LARGE, based on speed requirements and matrix size. (GEMM was faster but could not calculate larger matrices.) With this release, the GEMM procedure has been enhanced and GEMM_LARGE is no longer required.
Changes to nzMatrix New Random Number Generators This Netezza Analytics release introduces a new set of wrappers on the Intel Math Kernel Library® random number generators (RNGs). The API provides a set of stored procedures that generate matrices filled with random values.
Changes to Netezza Spatial Changes to Netezza Spatial Spatial precision has changed such that coordinate values display only the value's significant digits up to fifteen digits of precision which is the maximum for 64-bit floating point values. Prior to 2.0, by default, the user would always see 16 decimal digits. This means that any trailing 0's at the end of a value will now be truncated in this release.
Issues Fixed In Release 2.0 Reference Topic/Area Issue Description EXT-1084 PCA Performance Improvements EXT-1509 ARULE (formerly FPGROWTH) Performance Improvements EXT-1518 Netezza Matrix Engine When using RCV2SIMPLE_NUM or RCV2SIMPLE to convert a row/column/value table to a “simple” matrix table may fail if the number of projected columns is greater than 1600. EXT-1591 DECTREE Performance Improvements for large datasets.
Known Issues in Release 2.0 The following are known issues in Release 2.0. Those references numbers shown in red have been fixed in a later patch release. Reference Topic/Area Netezza Matrix Engine Issue Description / Workaround Using CTRL-C in nzsql typically aborts and rolls back the transaction in progress. However, it is possible that the Matrix Engine processes continue running and consuming resources. To check if any Matrix Engine processes are running, use the following SQL query: CALL NZA..
Known Issues in Release 2.0 Reference EXT-1107 Topic/Area Netezza Spatial Package Issue Description / Workaround A ST_DWithin function performed on two points does not return TRUE when increasing the distance value past 18945535.
Known Issues in Release 2.0 Reference Topic/Area Issue Description / Workaround EXT-1413 Netezza Analytics Some objects in the database are owned by ADMIN instead of INZAUSER, which may cause access issues. EXT-1593 DECTREE, REGTREE Tree scoring has a non linear scoring curve.
Known Issues in Release 2.0 Reference EXT-2209 and EXT-2101 Topic/Area Metadata Management Issue Description / Workaround nzconvertsyscase does not work in conjunction with Netezza Analytics. Conversion of the system case from uppercase to lowercase or vice versa using the command "nzconvertsyscase" does not convert the metadata management tables. Before you run this command, you must (for all databases) drop all analytics models and use the nza..cleanup() procedure to remove the metadata.
Known Issues in Release 2.0 Reference Topic/Area Issue Description / Workaround EXT-2319 nzSpatial ST_PointOnSurface incorrectly returning a point with empty polygons EXT-2328 Bayesian Networks Bayesian Networks are not deterministic in the choice of VARX and VARY EXT-2332 KMEANS The number of numeric columns supported are limited to 55 in this release when Mahalanobis distance is used for K-means clustering.
Netezza Analytics Release 2.0.1 The Netezza Analytics Release 2.0.1 patch release contains bug fixes and improvements to the documentation. Call Interface Changes The call interface to the following analytic function has changed in this release: Prior Release (old way) (Sample partial call) call nza..BITABLE( 'incolumn1=income; incolumn2=education ...') Release 2.0 (new way) (Sample partial call) call nza..BITABLE(' 'incolumn=income:x; education:y, …') Backward Compatible? No Issues Fixed In Release 2.
Documentation Changes Documentation Changes Release 2.0.1 contains the following two manuals documenting the Netezza Analytics map/reduce functionality.
Notices and Trademarks Notices This information was developed for products and services offered in the U.S.A. IBM may not offer the products, services, or features discussed in this document in other countries. Consult your local IBM representative for information on the products and services currently available in your area. Any reference to an IBM product, program, or service is not intended to state or imply that only that IBM product, program, or service may be used.
Notices and Trademarks IBM Corporation 26 Forest Street Marlborough, MA 01752 U.S.A. Such information may be available, subject to appropriate terms and conditions, including in some cases, payment of a fee. The licensed program described in this document and all licensed material available for it are provided by IBM under terms of the IBM Customer Agreement, IBM International Program License Agreement or any equivalent agreement between us.
Notices and Trademarks mon law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at “Copyright and trademark information” at ibm.com/legal/copytrade.shtml. The following terms are trademarks or registered trademarks of other companies: Adobe is a registered trademark of Adobe Systems Incorporated in the United States, and/or other countries.
Notices and Trademarks environment. This equipment generates, uses, and can radiate radio-frequency energy and, if not installed and used in accordance with the instruction manual, may cause harmful interference to radio communications. Operation of this equipment in a residential area is likely to cause harmful interference, in which case users will be required to correct the interference at their own expense.