IP Library Granted Patent US 7,124,240
Granted Patent B2
US 7,124,240 · App. 10/782,386 · Granted Oct 17, 2006

Method and circuit for increasing the memory access speed of an enhanced synchronous SDRAM

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 7,124,240
App. No.
10/782,386
Granted
Oct 17, 2006
Kind
B2
Abstract

An SDRAM and method for operating it provide for increased data access speed. The SDRAM includes a central memory region with memory blocks arranged in sets on respective opposite sides. A plurality of primary sense amplifier sets are provided, each set being associated with a respective pair of the memory blocks and located adjacent thereto. A row cache is provided in the central memory region, and row decoders decode a row address in response to a “bank activate” command and move data from a decoded row address into a primary sense amplifier set associated with a memory block containing the decoded row address and into the row cache, prior to application of a “read” command to the SDRAM. Column decoders decode a column address in response to a “read” command and for reading data from the cache in accordance with the decoded column address.

Claims (21)

1. A method for reading data from a synchronous memory of the type having data cells arranged in rows and columns, comprising:

arranging said synchronous memory in a symmetrical layout along a horizontal direction to include:

a left plurality of N memory portions including a left memory block, a central sense amplifier block, and a right memory block arranged along said horizontal direction;

one centrally located single row cache;

a secondary sense amplifier block co-located with said centrally located single row cache; and

a right plurality of N memory portions including a left memory block, a central sense amplifier block, and a right memory block arranged along said horizontal direction, said single row cache disposed between said left plurality of memory portions and said right plurality of memory portions,

wherein N is at least equal to two;

receiving an initial command that is not a “read” command and receiving row address data for reading contents of a row of said synchronous memory selected by said row address data, said initial command exclusive of colunm address data;

moving said contents of said row into said single row cache;

after said contents of said row have been moved into said single row cache, receiving a “read” command and column address data; and

in response to said “read” command, reading data from said single row cache at a column address specified by said column address data for output by said synchronous memory.

2. The method of claim 1 wherein said initial command is received substantially concurrently with said row address data.

3. The method of claim 1 wherein said “read” command and said column address data are received substantially concurrently.

4. The method of claim 1 further comprising moving data read from said single row cache to an output of said synchronous memory after a predetermined number of clock cycles after said “read” command.

5. The method of claim 4 wherein moving said data read from said row cache to an output of said synchronous memory after a predetermined number of clock cycles comprises moving said data read from said single row cache to an output of said memory after two clock cycles.

6. The method of claim 4 wherein said predetermined number of clock cycles is two.

7. The method of claim 1 wherein said receiving an initial command comprises receiving a “bank activate” command.

8. The method of claim 1 further comprising performing a first precharging operation prior to receiving said initial command.

9. The method of claim 4 further comprising initiating a memory operation after said contents of said row have been moved into said single row cache and before said data read from said single row cache has been moved to said output of said synchronous memory.

10. The method of claim 9 wherein said memory operation comprises a precharging operation.

11. The method of claim 9 wherein said synchronous memory comprises a SDRAM array.

Assignments (3)
MERGER Recorded Dec 7, 2010
From: PURPLE MOUNTAIN SERVER LLC
To: INTELLECTUAL VENTURES I LLC
Reel/Frame 025467/0102 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 13, 2007
From: PETERS, MICHAEL
To: ENHANCED MEMORY SYSTEMS, INC.
Reel/Frame 020104/0241 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded May 24, 2004
From: RAMTRON INTERNATIONAL CORPORATION; ENHANCED MEMORY SYSTEMS, INC.
To: PURPLE MOUNTAIN SERVER LLC
Reel/Frame 014662/0505 →