IP Library Granted Patent US 9,171,593
Granted Patent B2
US 9,171,593 · App. 14/117,295 · Granted Oct 27, 2015

Multi-granularity parallel storage system

Inventors: Donglin Wang (Beijing, CN); Zijun Liu (Beijing, CN); Xiaojun Xue (Beijing, CN); Xing Zhang (Beijing, CN); Zhiwei Zhang (Beijing, CN); Shaolin Xie (Beijing, CN)
Assignee: Institute of Automation, Chinese Academy of Sciences
G11C7/1072G06F3/0601G06F9/3895G06F12/02G06F12/0607G11C11/40615G11C21/00
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,171,593
App. No.
14/117,295
Granted
Oct 27, 2015
Kind
B2
Abstract

A multi-granularity parallel storage system including a plurality of memories, a shift generator, an address increment lookup unit, an address shifter, a row address generator, and a plurality of address adders. The shift generator is configured to generate a shift value. The address increment lookup unit is configured to generate input data for the address shifter. The address shifter is configured to cyclically shift the input data rightward by Shift elements and then output the shifted data. The row address generator is configured to generate a row address RowAddr and input the generated row address RowAddr to the other input terminal of each address adder. Each address adder is configured to perform a non-sign addition of the input data at the two input terminals to obtain a read/write (R/W) address for one of the memories and input the R/W address to an address input terminal of the memory.

Claims (17)

1. A multi-granularity parallel storage system comprising:

a plurality of memories;

a shift generator configured to generate a shift value Shift based on an externally-input address Addr;

an address increment lookup unit configured to generate input data based on a read/write (R/W) granularity g;

an address shifter configured to cyclically shift the input data rightward by Shift elements and output the shifted data;

a row address generator configured to generate a row address RowAddr based on an externally-input address Addr; and

a plurality of address adders each configured to perform a non-sign addition of the shifted data received at one input terminal and the generated row address RowAddr received at another input terminal to obtain a R/W address for one of the memories, and input the R/W address to an address input terminal of the memory.

2. The storage system of claim 1 , comprising:

a write shifter configured to cyclically shift externally-input written data rightward by Shift elements, and output the shifted data to the plurality of memories, and

a read shifter configured to receive data from the plurality of memories, cyclically shift the data leftward by Shift elements, and output the shifted data externally.

3. The storage system of claim 2 , wherein the address shifter, the write shifter, and the read shifter each have a shift control terminal; the output terminal of the shift generator is coupled to each of the shift control terminals of the address shifter, the write shifter, and the read shifter, so as to input the shift value Shift to the address shifter, the write shifter, and the read shifter.

4. The storage system of claim 2 , wherein the address shifter, the write shifter, and the read shifter each comprises a multiplexer.

5. The storage system of claim 1 , wherein the bit width of the storage system is W, K=log 2 W, and the memories support K+1 R/W granularities; and the address increment lookup unit comprises a plurality of columns, each corresponding to one of the R/W granularities.

6. The storage system of claim 5 , wherein the plurality of memories constitute a plurality of memory blocks, each having a size of N; and the jth element from bottom to top in the ith column from left to right in the address increment lookup unit has a value of [(j/2 i )×2 i ]×D, where “/” represent division operation with no remainder, D=N/W, 0≦i≦K, 0≦j<W.

7. The storage system of claim 6 , wherein the shift generator is further configured to generate the value Shift according to the following formula: Shift=(Addr % W)+(Addr/N), where “%” represents modulo operation, and “/” represent division operation with no remainder.

8. The storage system of claim 6 , wherein the row address generator is further configured to generate the row address RowAddr according to the formula RowAddr=Addr/W, where “/” represent division operation with no remainder.

9. The storage system of claim 1 , wherein each memory has a write enable terminal to which a write enable signal WE is directly inputted.

Assignments (4)
CHANGE OF NAME Recorded Jun 12, 2026
From: SHANGHAI SMARTLOGIC TECHNOLOGY LTD.
To: SHANGHAI SMART LOGIC TECHNOLOGY CO., LTD.
Reel/Frame 074931/0423 →
CHANGE OF NAME AND ADDRESS Recorded Apr 17, 2023
From: BEIJING SMARTLOGIC TECHNOLOGY LTD.
To: SHANGHAI SMARTLOGIC TECHNOLOGY LTD.
Reel/Frame 063348/0748 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Dec 10, 2017
From: INSTITUTE OF AUTOMATION, CHINESE ACADEMY OF SCIENCES
To: BEIJING SMARTLOGIC TECHNOLOGY LTD.
Reel/Frame 044345/0288 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 13, 2013
From: WANG, DONGLIN; LIU, ZIJUN; XUE, XIAOJUN; ZHANG, XING; ZHANG, ZHIWEI; XIE, SHAOLIN
To: INSTITUTE OF AUTOMATION, CHINESE ACADEMY OF SCIENCES
Reel/Frame 031595/0004 →
Continuity (1)
Related Publication 20140344515A1 · Nov 20, 2014