Scalable High-Bandwidth Architecture for Lossless Compression
First Claim
Patent Images
1. A system comprising:
- hardware logic circuitry to perform data compression by;
generating a multiple hash table that comprises at least a first hash table that includes latest positions for hash indexes and a second hash table that includes second latest positions for hash indices;
reading hash values from the first hash table and the second hash table simultaneously at a first clock rate; and
routing the read hash values to respective banks that operate at a second clock rate that is different from the first clock rate.
1 Assignment
0 Petitions
Accused Products
Abstract
A method of lossless data compression includes receiving a set of parallel data strings; determining compression hash values for each of the parallel data strings; determining bit matches among portions of each of the parallel data strings based, at least in part, on the compression hash values; selecting among literals and the bit matches for each of the parallel data strings; and applying Huffman encoding to the selected literals or the selected bit matches.
-
Citations
21 Claims
-
1. A system comprising:
hardware logic circuitry to perform data compression by; generating a multiple hash table that comprises at least a first hash table that includes latest positions for hash indexes and a second hash table that includes second latest positions for hash indices; reading hash values from the first hash table and the second hash table simultaneously at a first clock rate; and routing the read hash values to respective banks that operate at a second clock rate that is different from the first clock rate. - View Dependent Claims (2, 4, 5, 6, 7, 21)
-
3. (canceled)
-
8. A computing device comprising:
a hardware data compression pipeline accelerator including; a hash calculation module to receive a set of parallel data strings and to determine hash indexes for each of the parallel data strings; a hash table update module to read latest positions for each hash index and update the read latest positions with current string positions; a string match module to determine matches among portions of each of the parallel data strings based, at least in part, on the read latest positions; and a match selection module to select among literals and the matches for each of the parallel data strings. - View Dependent Claims (9, 10, 11, 12, 13, 14, 15)
-
16. A computing device comprising:
-
a memory device to store data; and a hardware data compression pipeline including; a string match module to determine bit matches among positions of each of a set of parallel data strings of the data; and a match selection module to choose among the bit matches that will be used to encode the data. - View Dependent Claims (17, 18, 19, 20)
-
Specification