Accuracy-conserving floating-point value aggregation

A computer system for generating data structures for computer memory and storing generated data therein, the computer system comprising:

• one or more computer processors;
• one or more computer readable storage medium;
• program instructions stored on the computer readable storage medium for execution by at least one of the one or more processors, the program instructions comprising:
• program instructions to generate a plurality of provisional data structures that respectively have (i) a significand with an initial value of zero and (ii) an exponent that is included in a range of exponents;
• program instructions to generate a test data structure from a first provisional data structure of the plurality of provisional data structures by modifying a significand of the first provisional data structure using a significand of a given floating-point number included in a plurality of floating point numbers such that the significand of the first provisional data structure has a non-zero value;
• program instructions to modify at least one provisional data structure of the plurality of provisional data structures based on a content of the test data structure;
• and
• program instructions to generate a new data structure that represents respective contents of the plurality of provisional-data structures, wherein the plurality of provisional data structures includes at least one provisional data structure that was modified to include respective significands of at least two floating-point numbers included in the plurality of floating point numbers, wherein the new data structure represents a value that is expressible by a limited number of memory bits.

A computer program product for generating data structures for computer memory and storing generated data therein, the computer program product comprising:

• one or more computer-readable storage media and program instructions stored on the one or more computer-readable storage media, the program instructions comprising:
• program instructions to generate a plurality of provisional data structures that respectively have (i) a significand with an initial value of zero and (ii) an exponent that is included in a range of exponents;
• program instructions to generate a test data structure from a first provisional data structure of the plurality of provisional data structures by modifying a significand of the first provisional data structure using a significand of a given floating-point number included in a plurality of floating point numbers such that the significand of the first provisional data structure has a non-zero value;
• program instructions to modify at least one provisional data structure of the plurality of provisional data structures based on a content of the test data structure;
• and
• program instructions to generate a new data structure that represents respective contents of the plurality of provisional-data structures, wherein the plurality of provisional data structures includes at least one provisional data structure that was modified to include respective significands of at least two floating-point numbers included in the plurality of floating point numbers, wherein the new data structure represents a value that is expressible by a limited number of memory bits.