Hewlett Packard Enterprise Development LP

Hewlett-packard Development Company, L.p.

Martin Foltin

John Paul Strachan

Gregg Bernard Lesartre

Catherine Graves

Dejan S Milojicic

Amit S Sharma

Craig William Warner

Sai Rahul Chalamalasetti

Paolo Faraboschi

Yoocharn Jeon

Sergey Serebryakov

Sean Tyler

Brian Foutz

Naveen Muralimanohar

Suhas Kumar

Chris Michael Brueggen

Aalap Tripathy

Erik Ordentlich

Brent Buchanan

Harvey Ray

James Ignowski

Harvey White, Jr

Eun Sub Lee

Richard Stanley Williams

Gary A Gibson

Suparna Bhattacharya

Richard H Henze

Kevin L Miller

Emmanuelle J Merced Grafals

Greg Astfalk

Clinton Harold Parker

Brian S Birk

Annmary Justine Koomthanam

Jeffrey Alan Lucas

Luke Whitaker

William Edward White

Rodrigo Jose Da Rosa Antunes

Francisco Plínio Oliveira Silveira

Joaquim Gomes Da Costa Eulalio De Souza

Jefferson Rodrigo Alves Cavalcante

Lucas Leite

Arthur Carvalho Walraven Da Cunha

Monycky Vasconcelos Frazao

Alex Ferreira Ramires Trajano

Glaucimar Da Silva Aguiar

Jacquelyn M Ingemi

Systems and methods are provide for automatically constructing data lineage representations for distributed data processing pipelines. These data lineage representations (which are constructed and stored in a central repository shared by the multiple data processing sites) can be used to among other things, clone the distributed data processing pipeline for quality assurance or debugging purposes. Examples of the presently disclosed technology are able to construct data lineage representations for distributed data processing pipelines by (1) generating a hash content value for universally identifying each data artifact of the distributed data processing pipeline across the multiple processing stages/processing sites of the distributed data processing pipeline; and (2) creating an data processing pipeline abstraction hierarchy for associating each data artifact to input and output events for given executions of given data processing stages (performed by the multiple data processing sites).

Method to track and clone data artifacts associated with distributed data processing pipelines

Systems and methods are provided to improve traditional chip processing. Using crossbar computations, the convolution layer can be flattened into vectors, and the vectors can be grouped into a matrix where each row or column is a flattened filter. Each submatrix of the input corresponding to a position of a convolution window is also flattened into a vector. The convolution is computed as the dot product of each input vector and the filter matrix. Using intra-crossbar computations, the unused space of the crossbars is used to store replicas of the filters matrices and the unused space in XIN is used to store more elements of the input. In inter-crossbar computations, the unused crossbars are used to store replicas of the filters matrices and the unused XINs are used to store more elements of the input. Then, the method performs multiple convolution iterations in a single step.

System and method for processing convolutions on crossbar-based neural network accelerators for increased inference throughput

A crossbar array includes a number of memory elements. An analog-to-digital converter (ADC) is electronically coupled to the vector output register. A digital-to-analog converter (DAC) is electronically coupled to the vector input register. A processor is electronically coupled to the ADC and to the DAC. The processor may be configured to determine whether division of input vector data by output vector data from the crossbar array is within a threshold value, and if not within the threshold value, determine changed data values as between the output vector data and the input vector data, and write the changed data values to the memory elements of the crossbar array.

Acceleration of model/weight programming in memristor crossbar arrays

In some examples, a device includes a first processing core comprising a resistive memory array to perform an analog computation, and a digital processing core comprising a digital memory programmable with different values to perform different computations responsive to respective different conditions. The device further includes a controller to selectively apply input data to the first processing core and the digital processing core.

Resistive and digital processing cores

Systems and methods are provided for enabling coexistence of Information Technology (IT) systems and Operational Technology (OT) systems, where advanced computing functionality realized by the IT systems can be applied to legacy applications and incumbent hardware technologies resident in the OT systems. A distributed control node (DCN) implemented between the IT and OT systems may comprise a microcontroller system partitioned into two processor clusters. Microservices associated with the IT systems are provisioned to a high performance processor cluster, and legacy applications running bare metal associated with the OT systems are provisioned to a real-time processor cluster. Partitioning of the microcontroller system allows for interoperability between the microservices and the legacy applications.

Scalable microservices-driven industrial IoT controller architecture

Systems and methods are provided for a multi-die dot-product engine (DPE) to provision large-scale machine learning inference applications. The multi-die DPE leverages a multi-chip architecture. For example, a multi-chip interface can include a plurality of DPE chips, where each DPE chip performs inference computations for performing deep learning operations. A hardware interface between a memory of a host computer and the plurality of DPE chips communicatively connects the plurality of DPE chips to the memory of the host computer system during an inference operation such that the deep learning operations are spanned across the plurality of DPE chips. Due to the multi-die architecture, multiple silicon devices are allowed to be used for inference, thereby enabling power-efficient inference for large-scale machine learning applications and complex deep neural networks. The multi-die DPE can be used to build a multi-device DNN inference system performing specific applications, such as object recognition, with high accuracy.

Multi-die dot-product engine to provision large scale machine learning inference applications

A DPE memristor crossbar array system includes a plurality of partitioned memristor crossbar arrays. Each of the plurality of partitioned memristor crossbar arrays includes a primary memristor crossbar array and a redundant memristor crossbar array. The redundant memristor crossbar array includes values that are mathematically related to values within the primary memristor crossbar array. In addition, the plurality of partitioned memristor crossbar arrays includes a block of shared analog circuits coupled to the plurality of partitioned memristor crossbar arrays. The block of shared analog circuits is to determine a dot product value of voltage values generated by at least one partitioned memristor crossbar array of the plurality of partitioned memristor crossbar arrays.

Self-healing dot-product engine

Examples described herein relate to a security system consistent with the disclosure. For instance, the security system may comprise a sensor interface bridge connecting a gateway to an input/output (I/O) card, a Field Programmable Gate Array (FPGA) to scan data to detect an anomaly in the data while the data is in the sensor interface bridge, where a learning neural network accelerator Application-Specific Integrated Circuit (ASIC) is integrated with the FPGA and send the data without an anomaly to the gateway.

Security and anomaly detection for Internet-of-Things devices

Disclosed techniques provide for dynamically changing precision of a multi-stage compute process. For example, changing neural network (NN) parameters on a per-layer basis depending on properties of incoming data streams and per-layer performance of an NN among other considerations. NNs include multiple layers that may each be calculated with a different degree of accuracy and therefore, compute resource overhead (e.g., memory, processor resources, etc.). NNs are usually trained with 32-bit or 16-bit floating-point numbers. Once trained, an NN may be deployed in production. One approach to reduce compute overhead is to reduce parameter precision of NNs to 16 or 8 for deployment. The conversion to an acceptable lower precision is usually determined manually before deployment and precision levels are fixed while deployed. Disclosed techniques and implementations address automatic rather than manual determination or precision levels for different stages and dynamically adjusting precision for each stage at run-time.

Adjustable precision for multi-stage compute processes

Devices and methods are provided. In one aspect, a device for driving a memristor array includes a substrate including a well having a bottom layer, a first wall and a second wall. The substrate is formed of a strained layer of a first semiconductor material. A vertical JFET is formed in the well. The vertical JFET includes a vertical gate region formed in a middle portion of the well with a gate region height less than a depth of the well. A channel region is formed of an epitaxial layer of a second semiconductor wrapped around the vertical gate region. Vertical source regions are formed on both sides of a first end of the vertical gate region, and vertical drain regions are formed on both sides of a second end of the vertical gate region.

Vertical JFET device for memristor array interface

A computer system includes multiple memory array components that include respective analog memory arrays which are sequenced to implement a multi-layer process. An error array data structure is obtained for at least a first memory array component, and from which a determination is made as to whether individual nodes (or cells) of the error array data structure are significant. A determination can be made as to any remedial operations that can be performed to mitigate errors of significance.

Determining significance levels of error values in processes that include multiple layers

Systems and methods for providing write process optimization for memristors are described. Write process optimization circuitry manipulates the memristor's write operation, allowing the number of cycles in the write process is reduced. Write process optimization circuitry can include write current integration circuitry that measures an integral of a write current over time. The write optimization circuitry can also include shaping circuitry. The shaping circuitry can shape a write pulse, by determining the pulse's termination, width, and slope. The write pulse is shaped depending upon whether the target memristor device exhibits characteristics of 'maladroit' cells or 'adroit' cells. The pulse shaping circuitry uses the integral and measured write current to terminate the write pulse in a manner that allows the memristor, wherein having maladroit cells and adroit cells, to reach a target state. Thus, utility of memristors is enhanced by realizing an optimized write process with decrease latency and improved efficiency.

Methods and systems for highly optimized memristor write process

In various examples, a device comprises a memory. The memory comprises a plurality of dies and logic. The logic may: determine a tolerable bit error rate (BER) of the memory based on whether one of the plurality of dies has failed, and adjust a parameter of the memory based on the tolerable BER.

Adjusting memory parameters

Growing community of inventors

Fort Collins, CO, United States of America

Martin Foltin