发明名称 NON-DETERMINISTIC FINITE AUTOMATON OVERFLOW RECOVERY
摘要 Disclosed is a method of recovering from overflow of a hardware dynamically reconfigurable NFA cell array, to find matches within a symbol stream to regular expression or similar rules without missing matches due to overflow. Upon overflow, active states are selected to spill from the cell array, saving state information and spill position. Scanning continues a limited distance, with additional overflow spills possible, to a selected end of segment position where all active end states are removed and recorded. A re-scan of the segment from the first overflow position begins with each previously spilled state re-injected at the same position it was spilled from. At the end of the segment, saved end states are re-injected and scanning continues. RE-scans may iterate if there was additional overflow. NFA states may be assigned color codes, with connected states receiving the same color, to aid in efficient overflow recovery.
申请公布号 US2014164309(A1) 申请公布日期 2014.06.12
申请号 US201213710582 申请日期 2012.12.11
申请人 LSI CORPORATION 发明人 Ruehle Michael;Khinvasara Rahul Indarbhan
分类号 G06N5/02 主分类号 G06N5/02
代理机构 代理人
主权项 1. A method of recovering from an overflow in an NFA in a dynamically configurable NFA cell array, comprising: scanning an input stream; recognizing an overflow condition in the NFA cell array; ejecting an overflow set of active states from the cell array into one of at least one state buffer, said set of active states comprising a portion of the active states in the NFA cell array; determining a segment length in the input stream from a current position to a first end of segment position; continuing scanning of the input symbols to the first end of segment position; ejecting an end of scan set of active states into one of the at least one state buffer, said end of scan set of active states comprising all remaining active states in the NFA cell array at the end of the step of continuing scanning; and rescanning the segment.
地址 Milpitas CA US
您可能感兴趣的专利