Kepler Computing Inc.

Debo Olaosebikan

Sasikanth Manipatruni

Amrita Mathuriya

Rajeev Kumar Dokania

Tanay Gosavi

Noriyuki Sato

Pratyush Pandey

Christopher B Wilkerson

Rafael Rios

Ikenna Odinaka

Niloy Mukherjee

Somilkumar J Rathi

James David Clarkson

Mustansir Yunus Mukadam

Pramod Kolar

Biswajeet Guha

Mauricio Manfrini

Darshak Doshi

Erik Unterborn

Ahmad Tavakoli

Described herein is a read and write scheme to improve memory reliability. In at least one embodiment, one or more circuitries are provided to perform logic 0 write operation in a first phase and logic 1 write operation in a second phase for a plurality of bit-cells controlled by a word-line. In at least one embodiment, an individual bit-cell comprises a transistor having a gate terminal coupled to the word-line; and a capacitor including non-linear polar material, wherein the capacitor has a first terminal coupled to a plate-line and a second terminal coupled to the transistor, wherein a source or drain terminal of the transistor is coupled to a bit-line.

Time decoupled write operations for non-linear polar material based memory

Described herein is a memory sensing scheme that improves noise margin. In at least one embodiment, one or more circuitries are described that are coupled to a bit-cell, wherein the bit-cell is coupled to a plate-line and a bit-line, wherein the one or more circuitries are to sense a bit-value stored in the bit-cell based, at least in part, on a first floating charge on a plate-line and a second floating charge on a bit-line.

Apparatus and method to improve sensing noise margin in a non-linear polar material based bit-cell

Described herein is a memory bit-cell that results in lower leakage and higher sensing margin. In at least one embodiment, a memory bit-cell comprises a plurality of capacitors, wherein an individual capacitor is coupled to a node and an individual plate-line. In at least one embodiment, memory bit-cell comprises a first transistor coupled to the node. In at least one embodiment, memory bit-cell comprises a second transistor coupled in series with the first transistor, wherein the second transistor is coupled to a bit-line, wherein the first transistor or the second transistor is controllable by a word-line, and wherein the word-line is parallel to the individual plate-line.

Non-linear polar material based multi-capacitor high density bit-cell

An apparatus and configuring scheme where a capacitive input circuit can be programmed to perform different logic functions by adjusting the switching threshold of the capacitive input circuit. Digital inputs are received by respective capacitors on first terminals of those capacitors. The second terminals of the capacitors are connected to a summing node. A pull-up and pull-down device are coupled to the summing node. The pull-up and pull-down devices are controlled separately. During a reset phase, the pull-up and/or pull-down devices are turned on or off in a sequence, and inputs to the capacitors are set to condition the voltage on node n. As such, a threshold for the capacitive input circuit is set. After the reset phase, an evaluation phase follows. In the evaluation phase, the output of the capacitive input circuit is determined based on the inputs and the logic function configured during the reset phase.

Stacked non-planar capacitors based multi-function linear threshold gate with input based adaptive threshold

An apparatus and configuring scheme where a paraelectric capacitive input circuit can be programmed to perform different logic functions by adjusting the switching threshold of the paraelectric capacitive input circuit. Digital inputs are received by respective capacitors on first terminals of those capacitors. The second terminals of the capacitors are connected to a summing node. A pull-up and pull-down device are coupled to the summing node. The pull-up and pull-down devices are controlled separately. During a reset phase, the pull-up and/or pull-down devices are turned on or off in a sequence, and inputs to the capacitors are set to condition the voltage on node n1. As such, a threshold for the capacitive input circuit is set. After the reset phase, an evaluation phase follows. In the evaluation phase, the output of the capacitive input circuit is determined based on the inputs and the logic function configured during the reset phase.

Multi-function threshold gate with input based adaptive threshold and with stacked non-planar paraelectric capacitors

Described is a packaging technology to improve performance of an AI processing system. An IC package is provided which comprises: a substrate; a first die on the substrate, and a second die stacked over the first die. The first die includes memory and the second die includes computational logic. The first die comprises DRAM having bit-cells. The memory of the first die may store input data and weight factors. The computational logic of the second die is coupled to the memory of the first die. In one example, the second die is an inference die that applies fixed weights for a trained model to an input data to generate an output. In one example, the second die is a training die that enables learning of the weights. Ultra high-bandwidth is changed by placing the first die below the second die. The two dies are wafer-to-wafer bonded or coupled via micro-bumps.

Ultra high-bandwidth artificial intelligence (AI) processor with DRAM under the processor

A packaging technology to improve performance of an AI processing system resulting in an ultra-high bandwidth system. An IC package is provided which comprises: a substrate; a first die on the substrate, and a second die stacked over the first die. The first die can be a first logic die (e.g., a compute chip, CPU, GPU, etc.) while the second die can be a compute chiplet comprising ferroelectric or paraelectric logic. Both dies can include ferroelectric or paraelectric logic. The ferroelectric/paraelectric logic may include AND gates, OR gates, complex gates, majority, minority, and/or threshold gates, sequential logic, etc. The IC package can be in a 3D or 2.5D configuration that implements logic-on-logic stacking configuration. The 3D or 2.5D packaging configurations have chips or chiplets designed to have time distributed or spatially distributed processing. The logic of chips or chiplets is segregated so that one chip in a 3D or 2.5D stacking arrangement is hot at a time.

3D stack of split graphics processing logic dies

Stacked planar capacitors based multi-function linear threshold gate with input based adaptive threshold

A configuration for efficiently placing a group of capacitors with one terminal connected to a common node is described. The capacitors are stacked and folded along the common node. In a stack and fold configuration, devices are stacked vertically (directly or with a horizontal offset) with one terminal of the devices being shared to a common node, and further the capacitors are placed along both sides of the common node. The common node is a point of fold. In one example, the devices are capacitors. N number of capacitors can be divided in L number of stack layers such that there are N/L capacitors in each stacked layer. The N/L capacitors are shorted together with an electrode (e.g., bottom electrode). The electrode can be metal, a conducting oxide, or a combination of a conducting oxide and a barrier material. The capacitors can be planar, non-planar or replaced by memory elements.

Multi-input threshold gate having stacked and folded non-planar capacitors

Method of forming an artificial intelligence processor with three-dimensional stacked memory

A method for monetizing ferroelectric process development is described. In at least one embodiment, the method comprises procuring a target material based on a model driven selection which is based on charge, mass and magnetic moment, and/or mass of the atomic constituents of the target material. The method further comprises applying the target material to a fabrication process to build a ferroelectric device. The method further comprises generating a notification indicative of procurement of the target material and application of the target material. The method further comprises electronically transmitting the notification to a customer, wherein the notification includes an invoice having a line item associated with a cost of the procuring of the target material and application of the target material.

Iterative monetization of precursor in process development of non-linear polar material and devices

Memory bit-cell with stacked and folded planar capacitors

One transistor and N memory element based memory bit-cell having stacked and folded planar memory elements with and without offset

A ferroelectric memory chiplet in a multi-dimensional packaging. The multi-dimensional packaging includes a first die comprising a switch and a first plurality of input-output transceivers. The multi-dimensional packaging includes a second die comprising a processor, wherein the second die includes a second plurality of input-output transceivers coupled to the first plurality of input-output transceivers. The multi-dimensional packaging includes a third die comprising a coherent cache or memory-side buffer, wherein the coherent cache or memory-side buffer comprises ferroelectric memory cells, wherein the coherent cache or memory-side buffer is coupled to the second die via I/Os. The dies are wafer-to-wafer bonded or coupled via micro-bumps, copper-to-copper hybrid bond, or wire bond, Flip-chip ball grid array routing, chip-on-wafer substrate, or embedded multi-die interconnect bridge.

Growing community of inventors

San Francisco, CA, United States of America

Debo Olaosebikan