HARDWARE COMPRESSION TO FIND BACKWARD REFERENCES WITH MULTI-LEVEL HASHES

摘要

Concurrently writing an uncompressed data element, if the uncompressed data element comprises an indication that it is valid, in a main hash table using a first address generated by a first hash function, and reading a first data element from the main hash table using the first address. Introducing a first pipeline delay for maintaining the uncompressed data element in a first data path until the first data element is read. Concurrently writing the first data element to a victim hash table, if the first data element comprises an indication that it is valid, using a second address generated by a second hash function, and reading a second data element from the victim hash table using a third address generated by the second hash function. Introducing a second pipeline delay for maintaining the uncompressed data element in the first data path until the second data element is read.

主权项

1. A method of processing an uncompressed data element received in a stream of data elements by a hardware logic system having a multi-stage pipeline structure and a distributed memory comprising a main hash table and at least one victim hash table, the method comprising:
a) concurrently writing the uncompressed data element, based on the uncompressed data element comprising an indication that it is a valid data element, in the main hash table using a first address generated by a first hash function from the uncompressed data element, and reading a first data element from the main hash table using the first address; b) introducing a first pipeline delay for at least one clock cycle for maintaining the uncompressed data element in a first data path until the first data element is read; c) concurrently writing the first data element to a victim hash table of the at least one victim hash table, based on the first data element comprising an indication that it is a valid data element, using a second address generated by a second hash function from the first data element, and reading a second data element from the victim hash table using a third address generated by the second hash function from the uncompressed data element; d) introducing a second pipeline delay for at least one clock cycle for maintaining the uncompressed data element in the first data path until the second data element is read; e) selecting a data element from at least the first and second data elements having a longest sequence that matches at least part of the uncompressed data element, based on the uncompressed data element and at least one of the first and second data elements comprising an indication that they are valid data elements; and f) using the longest sequence that matches for providing the uncompressed data element in a compressed format backward referring to the longest sequence that matches.