Sortable Floating-Point Format for Fast Bitwise Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The lack of hardware support for IEEE 754R standardization in microprocessors hinders efficient manipulation and comparison of floating-point numbers, leading to increased processing time and storage requirements, which is detrimental for database queries and data sort operations.
Innovation Solution
Converting floating-point numbers to a sortable floating-point format (SFP) that maintains the same bit length as IEEE 754R, allowing for bit-wise comparisons and compression, thereby enabling efficient processing and storage without requiring dedicated hardware support.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If floating-point numbers are stored in IEEE 754R format without hardware support, then the numbers can be represented in a standardized format, but comparison operations become computationally expensive and time-consuming
Solution Approach 1:
The patent creates a copy of the IEEE 754R formatted number in a sortable format that preserves the numerical value but reorganizes the bits to enable efficient comparison. This copy is used for comparison operations while the original IEEE 754R representation is maintained for standardized arithmetic operations, thus resolving the contradiction between maintaining standardization and enabling fast comparisons.
2Adaptability or versatility
If floating-point numbers are manipulated using software operations instead of hardware support, then compatibility with existing microprocessors is maintained, but processing speed and efficiency decrease
Solution Approach 1:
The patent changes the parameter representation of floating-point numbers by creating a sortable format that reorganizes the bits of IEEE 754R numbers. This format transformation enables hardware-efficient comparison and sort operations on existing microprocessors without requiring dedicated IEEE 754R hardware support, thus improving productivity while maintaining adaptability to existing processors.
3Reliability
If data is stored in a format that requires conversion for sorting operations, then the standardized format can be maintained, but additional processing steps and storage overhead are required
Solution Approach 1:
The patent creates a compact sortable format that is a direct bit-level transformation of IEEE 754R representation, serving as an efficient copy for sorting operations. This copy requires no additional storage overhead beyond what is already allocated for the floating-point number itself, and the transformation can be performed in-place or with minimal temporary storage, thus maintaining standardization while reducing processing complexity.
Data Source
AI summary
The invention comprises methods for manipulating floating point numbers on a microprocessor where the numbers are sortable. That is, the numbers obey lexicographical ordering. Hence, the numbers may be quickly compared using bit-wise comparison functions such as memcmp( ). Conversion may result in a sortable floating point number in the form of a sign, leading bits of the exponent, and sets of digit triples in the form of declets (sets of 10 bits). In a variable-length version, numbers may be compressed by storing the number of trailing zero declets in lieu of storing the zero declets themselves.