Back to results
US 2017/0249105 A1Published application
Systems and Methods for High Throughput Multi-Input Compression
AssigneeAvago Technologies General IP (Singapore) Pte. Ltd.
InventorsShaohua Yang, Han Fang, Wu Chang, Kelly Fitzpatrick
Life of this patent
Estimating the term…
Abstract
Systems and methods relating generally to data storage, and more particularly to systems and methods for encoding to modify the size of an information set.
Claims20
A compression based data encoding system, the system comprising:
- a FIFO having storage locations for N data portions, wherein N is an integer value, and wherein the FIFO is operable to:
- shift each of the N data portions by K positions within the FIFO on each clock cycle;
- receive K input data portions, wherein K is an integer value greater than one;
- and
- store the received K input data portions into the first K storage locations in the FIFO;
- a portion matching circuit operable to:
- compare a first input data portion with each of the N data portions in the FIFO to yield an input match set[1];
- for each of the K input data portions indicated as K[i] wherein i is an integer value between two and K, compare K[i] with each of K[1] to K[i−1] and the most recent N−(i−1) data portions in the FIFO to yield an input match set[i], wherein comparing the K[i] with each of K[1] to K[i−1] and the most recent N−(i−1) data portions in the FIFO is done in parallel with comparing the first input data portion with each of the N data portions in the FIFO;
- a segment matching circuit operable to:
- receive input match set[1] and input match set[i];
- identify a series of at least M preceding data portions within the N data portions of the FIFO that match a sequence of at least M data portions, wherein M is an integer value;
- and
- generate a code indicating a number of data portions in the at least M incoming data portions and a distance between the at least M incoming data portions and the at least M preceding data portions.
10 more elements
A compression based data encoding system, the system comprising:
- a FIFO operable to store a series of N data portions;
- a portion matching circuit operable to:
- compare a first input data portion with each of the N data portions in the FIFO to yield a first match set;
- compare a second input data portion with each of the first input data portion and the most recent N−1 data portions in the FIFO to yield a second match set, wherein comparing the second input data portion with each of the first input data portion and the most recent N−1 data portions in the FIFO is done in parallel with comparing the first input data portion with each of the N data portions in the FIFO;
- a segment matching circuit operable to:
- receive the first match set and the second match set;
- identify a series of at least M preceding data portions within the N data portions of the FIFO that match a sequence of at least M incoming data portions;
- and
- generate a code indicating a number of data portions in the at least M incoming data portions and a distance between the at least M incoming data portions and the at least M preceding data portions.
6 more elements
A method for compression based data encoding, the method comprising:
- providing a FIFO of a length N and operable to store N data portions;
- comparing a first input data portion with each of the N data portions in the FIFO to yield a first match set;
- comparing a second input data portion with each of the first input data portion and the most recent N−1 data portions in the FIFO to yield a second match set, wherein comparing the second input data portion with each of the first input data portion and the most recent N−1 data portions in the FIFO is done in parallel with comparing the first input data portion with each of the N data portions in the FIFO;
- identifying a series of at least M preceding data portions within the N data portions of the FIFO that match a sequence of at least M incoming data portions;
- and
- generating a code indicating a number of data portions in the at least M incoming data portions and a distance between the at least M incoming data portions and the at least M preceding data portions.
3 more elements
Dependencies are inferred from the claim text.