Seagate Technology Incorporated

Sk Hynix Nand Product Solutions Corp.

Intel Corporation

Cellport Systems, Inc.

Marc Tim Jones

Mark A Schmisseur

Michael T Klinglesmith

Bradley A Burres

Sivakumar Radhakrishnan

Eric Dahlen

Pankaj Kumar

Sin Sim Tan

Thomas M Slaight

Brian M Leitner

Geetani R Edirisooriya

David C Lee

Matthew R Wilcox

Kenneth C Haren

Saptarshi Mondal

Patrick J Kennedy

Charles W Spaur

Michael Tibbetts

Use of computational storage drives (CSDs) in a machine learning pipeline. The CSD may include a machine learning coprocessor capable of natively executing a machine learning model on raw data stored locally at the CSD. In turn, one or more lower order machine learning operations may be executed at a CSD in response to a read/transform command issued by a host. In turn, the CSD may return transformed data comprising data or metadata that is an output of the lower order machine learning operations. This approach may allow for application of a machine learning model locally to input data stored on the CSD having the machine learning coprocessor. This may avoid network bandwidth associated with traditional read and write operations for input and output data from a machine learning pipeline. Moreover, use of CSDs may provide highly parallelized processing using CSDs for application of machine learning operations at the edge of a network.

Deep learning computational storage drive

In accordance with embodiments disclosed herein, there are provided methods, systems, mechanisms, techniques, and apparatuses for presentation of direct accessed storage under a logical drive model; for implementing a distributed architecture for cooperative NVM Data protection; data mirroring for consistent SSD latency; for boosting a controller's performance and RAS with DIF support via concurrent RAID processing; for implementing arbitration and resource schemes of a doorbell mechanism, including doorbell arbitration for fairness and prevention of attack congestion; and for implementing multiple interrupt generation using a messaging unit and NTB in a controller through use of an interrupt coalescing scheme.

Presentation of direct accessed storage under a logical drive model

The technology disclosed herein pertains to a system and method for profiling performance of an embedded computation instruction set (CIS), the method including receiving a profiling component to a computational storage device (CSD), the profiling component being configured to measure one or more execution parameters of a computational instruction set (CIS), executing the CIS at a program slot in a computational storage processor of the CSD, monitoring the execution of the CIS to generate a log of the execution parameters of the CIS, and communicating the log to a host in response to receiving a get-log page command.

Embedded computation instruction performance profiling

The technology disclosed herein provides a method including determining one or more dedicated computations storage programs (CSPs) used in a target market for a computational storage device, storing the dedicated CSPs in one or more pre-programmed computing instruction set (CIS) slots in the computational storage device, translating one or more instructions of the dedicated CSPs for processing using a native processor, loading one or more instructions of programmable CSPs to a CSP processor implemented within an application specific integrated circuit (ASIC) of the computational storage device, and processing the one or more instructions of the programmable CSPs using the CSP processor.

Computational storage with pre-programmed slots using dedicated processor core

The technology disclosed herein provides a method including generating, at a PCIe interface, a functional capability table associating a plurality of computational functionalities (CFs) with a plurality of computational storage devices (CSDs) communicatively connected to each other via the PCIe interface, communicating the functional capability table to each of the plurality of CSDs, receiving, at a first of the plurality of CSDs (CSD), a request for a computational process including the plurality of CFs, determining that a first of the CFs (CF) is associated with the CSDand a second of the CFs (CF) is associated with a CSD, requesting a computational program memory (CPM) at the CSDand performing the CFat CSD

Computational pipelines for computational storage devices

A communications apparatus for enabling communications among networks, buses, devices and other subsystems having different communication requirements is provided. The communications apparatus offers an open architecture for interoperability among diverse subsystems. The communications apparatus is particularly applicable in a vehicle. In one embodiment, a portal apparatus located remote from the vehicle facilitates communications with the communications apparatus, particularly where the vehicle is part of a vehicle fleet. The communications apparatus includes a common module that has at least one protocol by which disparate subsystems can communicate with each other. One or more managers can be defined as being part of the common module. The disparate subsystem managers provide the necessary interfacing to enable the disparate subsystems to communicate between and among each other using the common module. Each such manager has dedicated functionalities to facilitate communications. Managers associated with the common module may include managers for use in configuration and diagnostic tasks, encryption/decryption, compression/de-compression, controlling data flow, security regulation, and selecting an acceptable link for outside communications. The common module also includes a base manager for registering disparate subsystem managers including associating a name with each of them.

Communications involving disparate protocol network/bus and device subsystems

Growing community of inventors

Longmont, CO, United States of America

Marc Tim Jones