Agere Systems Inc.

Lucent Technologies Inc.

At+t Corp.

At&t Bell Laboratories

Agere Systems Guardian Corp.

Philip W Diodato

Yiu-Huen Wong

Ruichen Liu

James Theodore Clemens

Chun-Ting Liu

Glenn B Alers

Hubert Rae McLellan, Jr

Girija J Narlikar

Stefanos Kaxiras

Harry T Weston

Girija Narlikar

A method and apparatus are disclosed for reducing leakage power in a cache memory. A cache decay technique is employed for both data and instruction caches that removes power from cache lines that have not been accessed for a predefined time interval, referred to as the decay interval. The cache-line granularity of the present invention permits a significant reduction in leakage power while at the same time preserving much of the performance of the cache. The decay interval is maintained using a timer that is reset each time the corresponding cache line is accessed. The decay interval may be fixed or variable. Once the decay interval timer exceeds a specified decay interval, power to the cache line is removed. Once power to the cache line is removed, the contents of the data and tag fields are allowed to decay and the valid bit associated with the cache line is reset. When a cache line is later accessed after being powered down by the present invention, a cache miss is incurred while the cache line is again powered up and the data is obtained from the next level of the memory hierarchy.

Method and apparatus for reducing leakage power in a cache memory by using a timer control signal that removes power to associated cache lines

An integrated circuit includes a semiconductor substrate with semiconductor devices formed therein and thereon, a first wiring layer located over the substrate, a second wiring layer located on the first wiring layer, and a capacitor. The capacitor has metal-based charge-storage electrodes that extend through the second wiring layer and at least part of the first wiring layer. The wiring layers have interconnect wire embedded therein.

Inter-wiring-layer capacitors

A process for fabricating trench capacitors in an interconnect layer of a semiconductor device is disclosed. In the process, at least one interconnect is formed in the interconnect layer, which is then planarized. To form the trench capacitor, a trench is formed in the dielectric material of the interconnect. The bottom of the trench communicates with a metal contact in the underlying layer. A barrier layer of material is formed on the interconnect layer and is anisotropically etched, leaving the barrier layer on the sidewalls of the trench. The lower plate of the capacitor is then formed by depositing a layer of metal over the interconnect layer. The layer of metal is then anisotropically etched, removing the metal on the surface of the interconnect layer and leaving the metal along the trench perimeter. The capacitor dielectric layer is then deposited over the interconnect layer and subsequently patterned. Another layer of barrier material is deposited on the interconnect layer. The metal for the top plate of the capacitor is then deposited over the interconnect layer. The top plate of the capacitor extends into the trench. The metal layer and barrier layer are then patterned to form the top plate of the capacitor.

Process for fabricating an integrated circuit device having capacitors with a multilevel metallization

An apparatus and method for constructing a temperature insensitive memory cell. This temperature insensitive memory cell operates as a static random access memory (SRAM) cell if a particular capacitor and transistor configuration is used. The temperature insensitive memory cell apparatus includes at least one transistor having a current leakage, and at least one capacitor electrically connected to the transistor. The capacitor acts as a load element for the memory cell. The capacitor has a temperature dependent capacitor leakage that tracks the current leakage of transistor as said at least one transistor as the transistor varies with temperature.

Temperature insensitive capacitor load memory cell

An apparatus and method for constructing a capacitor loaded memory cell. This capacitor loaded memory cell operates as a static random access memory (SRAM) cell if a particular capacitor and transistor configuration is used. Normally, capacitors are not an obvious choice as a load device for a memory cell because the intrinsic nature of capacitors is one that blocks the flow of direct current, the invention takes into account the secondary effects such as leakage of a particular dielectric used in the construction of the capacitor to modify the current blocking nature of the capacitor.

Capacitor loaded memory cell

In an electronic system such as an integrated circuit having a number of destination loads such as logic gates, signal is distributed along typically a zero'th level (e.g., polysilicon) electrical transmission line from an input terminal to the destination loads. The characteristics of the signal arriving at the destination loads are improved by (1) inserting an added electrical transmission line, and (2) connecting various nodes of the added electrical transmission line through auxiliary active devices, such as inverters, to various nodes on the zero'th level electrical transmission line. In one attractive arrangement, each of the auxiliary active devices has an electrical-current-drive capability that increases monotonically with the number of nodes intervening between it and the input terminal of the added electrical transmission line.

Circuitry for delivering a signal to different load elements located in

A digital logic circuit (100) is provided for multiplying, such as doubling, the frequency of an input clock pulse sequence of period T. The circuit in one embodiment includes complementarily clocked first and second chains of cascaded delay elements (12, 13 in A1, A2, A3, . . . and B1, B2, B3, . . . ). Further, the n'th one of set of clocked latches (14, 15, 16 in A2, A4, A6, . . . ) derives its input from the 2n'th one of the delay elements in the first chain, where n is a running integer index (n=1,2,3, . . . ). The circuit (100) also includes a set of two-input logic gates (11), one of whose inputs (IN) is the output (OU) of a separate one of the logic elements (12, 13) in the second chain and the other of whose inputs is an output (MO) of a separate one of the latches (14, 15, 16). Each of the outputs of these logic gates (11) is fed to a multiple input output logic gate (25) whose output has a desired double-frequency feature (edges at T/4) relative to the frequency of the clocked pulse sequence (CLK).

Growing community of inventors

Asbury, NJ, United States of America

Philip W Diodato