Static information storage and retrieval – Read/write circuit – Having particular data buffer or latch
Reexamination Certificate
2002-03-13
2004-03-16
Nguyen, Viet Q. (Department: 2818)
Static information storage and retrieval
Read/write circuit
Having particular data buffer or latch
C365S154000, C365S196000
Reexamination Certificate
active
06707721
ABSTRACT:
BACKGROUND OF INVENTION
A typical computer system has at least a microprocessor and memory. The microprocessor processes, i.e., executes, instructions to accomplish various tasks of the computer system. Such instructions, along with the data required by the microprocessor when executing these instructions, are stored in some form of memory.
FIG. 1
shows a typical computer system having a microprocessor (
10
) and some form of memory (
12
). The microprocessor (
10
) has, among other things, a central processing unit (also known and referred to as “CPU” or “execution unit”) (
14
) and a memory controller (also known as “load/store unit”) (
16
). The CPU (
14
) is where the actual arithmetic and logical operations of the computer system take place. To facilitate the execution of operations by the CPU (
14
), the memory controller (
16
) provides the CPU (
14
) with necessary instructions and data from the memory (
12
). The memory controller (
16
) also stores information generated by the CPU (
14
) into the memory (
12
).
Memory, as shown in
FIG. 1
, is typically formed by numerous storage cells, where each storage cell contains a bit of data. Memory organized in such a fashion is called a “memory array.” The data in each storage cell can have either a logic low value, i.e., ‘0,’ or a logic high value, i.e., ‘1.’ As a result, the value(s) of one or more storage cells are often used to represent numbers, characters, instructions, etc. Accordingly, to ensure that data in a storage cell is properly read and stored, important consideration must be afforded to how a memory array is designed.
FIG. 2
shows a typical implementation of a memory array (
20
). The memory array (
20
) has numerous storage cells (also known and referred to as “memory elements,” “memory cells,” and “register file cells”) (
36
), where each storage cell (
36
) is connected to a bit line (also known as “row line”) (
40
) and a word line (also known as “column line”) (
42
). Word lines (
42
) are selected using a row decoder (
32
) that drives a particular word line (
42
) using a word line driver (
34
). Bit lines (
30
) are selected using a column decoder (
30
), where a value on a bit line (
40
) is driven to a sense amplifier (
44
) by a bit line driver (
38
). A storage cell (
36
) is selected for a read/write operation by some combination of activating the bit line (
40
) connected to the storage cell (
36
) and activating the word line (
42
) connected to the storage cell (
36
).
Accurately and quickly reading data from a memory array, such as the one shown in
FIG. 2
, is challenging because of both large physical sizes of the memory array and a high number of entries, i.e., storage cells, in the memory array.
FIG. 3
shows a typical implementation of a storage cell (
36
). One technique used by designers to increase read operation performance is to use a pre-charge device (
50
) to pre-charge a bit line (
40
) to a logic high value, i.e., ‘1,’ prior to a read operation (also referred to as “pre-charge phase”), and then to perform, i.e., evaluate, the read operation (also referred to as “evaluation phase”) on the storage cell (
36
) by allowing the bit line (
40
) connected to the particular storage cell (
36
) to discharge if the particular storage cell (
36
) holds a ‘0’ value. If, however, the bit line (
40
) does not discharge, it implies that the particular storage cell (
36
) connected to the bit line holds a ‘1’ value.
Still referring to
FIG. 3
, during a read operation, a pass device (
48
) is switched ‘on’ by some input at its gate terminal. If the storage cell (
36
) holds a ‘0’ value, then a footer device (
46
), which has one terminal connected to ground, is ‘on’ causing the bit line (
40
) to discharge through the ‘on’ pass and footer devices (
48
,
46
).
In attempting to increase the speed of storage cell access, designers have increased the size of the pass device (shown in
FIG. 3
) to add drive strength to the storage cell. However, increasing the pass device size also increases the capacitance on the bit line to which the pass device is connected. Such an increase in bit line load slows down storage cell access. Thus, the net gain in speed due to increasing the pass device size is much smaller in proportion to the increase in the size of the pass device. Further still, the increase in size of the pass device leads to increased power consumption, which is in proportion to the increase in size of the pass device. Accordingly, an incremental gain in pass device size causes a corresponding increase in power consumption. Thus, there is a need for a memory design that has low power consumption without a loss in storage cell access speed.
SUMMARY OF THE INVENTION
According to one aspect of the present invention, an integrated circuit that has a memory comprises a bit line and a storage cell connected to the bit line, where the storage cell comprises a pass device connected to the bit line and a footer device connected to the pass device, where the footer device facilitates driving the bit line through the pass device based on a value stored in the storage cell, and where the pass device and the footer device are asymmetric.
According to another aspect, an integrated circuit comprises storing means for storing a value, propagating means for propagating the value from the storing means, a first driver means for driving the propagating means, and a second driving means for driving the propagating means, where the first driver means and the second driver means are asymmetric.
According to another aspect, a method for performing memory operations in an integrated circuit comprises: storing a value in a memory element, where the memory element is connected to a bit line; and facilitating driving the bit line through a pass device and a footer device based on the value stored in the memory element, where the pass device and the footer device are asymmetric.
Other aspects and advantages of the invention will be apparent from the following description and the appended claims.
REFERENCES:
patent: 5616948 (1997-04-01), Pfiester
patent: 5673230 (1997-09-01), Kuriyama
patent: 6141744 (2000-10-01), Wing So
patent: 6512712 (2003-01-01), Desai et al.
patent: 6522594 (2003-02-01), Scheuerlein
patent: 2002/0016579 (2002-02-01), Stenberg
Kant Shree
Ramachandran Aparna
Rao Miao
Singh Gajendra
Nguyen Viet Q.
Rosenthal & Osha L.L.P.
Sun Microsystems Inc.
LandOfFree
Low power memory design with asymmetric bit line driver does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Low power memory design with asymmetric bit line driver, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Low power memory design with asymmetric bit line driver will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3256298