Support for floating point "DECfloat" data types in DQM so floatingvalues not being rounded to a Double datatype in Cognos BI

Support for floating point "DECfloat" data types in DQM so floatingvalues not being rounded to a Double datatype
*************************************************************

The IEEE Standard IEEE 754-2008 (previously called IEEE 754r) for
binary floating-point arithmetic was revised to include the definition
of decimal floating-point arithmetic. This replaced the prior standard
IEEE-754-1985.

This is intended mainly to provide a robust reliable framework for
financial applications that are often subject to legal requirements
concerning rounding and precision of the results, because the binary
floating-point arithmetic may introduce small but unacceptable errors.

Vendors such as IBM have implemented hardware level support to
accelerate calculations as opposed to implementations in software (such
as DECIMAL (N,M)) typically involve.

Support for such a 'DECfloat' type must take into consideration the
following characteristics which a customer may assume IBM Cognos offers
once support is claimed:

1) which vendor data types are treated as a 'DECfloat' type (separate
stories should be added re DB2, Informix, Vertica, ... as appropriate)
2) what IBM Cognos has done to avoid automatically injecting
expressions (i.e cast as double, * 1.E0 etc) that could cause a type
conversion
3) how IBM Cognos allows a customer to control (Cognos instance,
package or model) what rounding mode Cognos BI defaults to
4) if IBM Cognos detects what rounding mode the RDBMS may be configured
what it does based on that fact (ignore, warn, ...)
5) if IBM Cognos supports SQL syntax extensions vendors may offer such
as IS [NOT] INFINITE, IS [NOT] NAN, SNAN
6) how automatic div-by-zero avoidance in Cognos BI can/should be used
when customer expects IEEE behaviour re 0 / 0, 1 / 0
7) if IBM Cognos supports quantize
8) how data is exported by IBM Cognos (save-as) or imported (personal
data) where someone presumes support for DECfloat
9) ensure documentation is provided with respect to customers updating
existing Cognos BI apps where those types may have been 'treated' as a
DOUBLE PRECISION etc

See attached documents and associated reference links:

Background references:

http://www.javaworld.com/article/2077257/learn-java/floating-point-
arithmetic.html

http://grouper.ieee.org/groups/754/

http://www.cs.berkeley.edu/wkahan/Mindless.pdf

Dense packed decimal
http://speleotrove.com/decimal/DPDecimal.html

Informix

http://publib.boulder.ibm.com/infocenter/idshelp/v10/index.jsp?
topic=/com.ibm.dapip.doc/dapip90.htm
http://www-01.ibm.com/support/docview.wss?uid=swg21615373

Java

http://www.ibm.com/developerworks/library/j-jtp0114/
http://docs.oracle.com/javase/7/docs/api/java/math/BigDecimal.html