Abstract: Technologies for dividing work across one or more accelerator devices include a compute device. The compute device is to determine a configuration of each of multiple accelerator devices of the compute device, receive a job to be accelerated from a requester device remote from the compute device, and divide the job into multiple tasks for a parallelization of the multiple tasks among the one or more accelerator devices, as a function of a job analysis of the job and the configuration of each accelerator device. The compute engine is further to schedule the tasks to the one or more accelerator devices based on the job analysis and execute the tasks on the one or more accelerator devices for the parallelization of the multiple tasks to obtain an output of the job.

Abstract: Gate-all-around structures having devices with source/drain-to-substrate electrical contact are described. An integrated circuit structure includes a first vertical arrangement of horizontal nanowires above a first fin. A first gate stack is over the first vertical arrangement of horizontal nanowires. A first pair of epitaxial source or drain structures is at first and second ends of the first vertical arrangement of horizontal nanowires. One or both of the first pair of epitaxial source or drain structures is directly electrically coupled to the first fin. A second vertical arrangement of horizontal nanowires is above a second fin. A second gate stack is over the second vertical arrangement of horizontal nanowires. A second pair of epitaxial source or drain structures is at first and second ends of the second vertical arrangement of horizontal nanowires. Both of the second pair of epitaxial source or drain structures is electrically isolated from the second fin.

Abstract: In one embodiment, a base die apparatus includes a conformal power delivery structure comprising a first electrically conductive layer defining one or more recesses, and a second electrically conductive layer at least partially within the recesses of the first electrically conductive layer and having a lower surface that generally conforms with the upper surface of the first electrically conductive layer. The conformal power delivery structure also includes a dielectric material between the surfaces of the first electrically conductive layer and the second electrically conductive layer that conform with one another. The conformal power delivery structure may be connected to connection pads of the base die apparatus, e.g., to provide power delivery to integrated circuit (IC) chips connected to the base die apparatus. The base die apparatus also includes bridge circuitry to connect IC chips with one another.

Abstract: Microelectronic assemblies, and related devices and methods, are disclosed herein. For example, in some embodiments, a microelectronic assembly may include a package substrate having a core substrate with a first conductive structure having a first thickness on the core substrate, and a second conductive structure having a second thickness on the core substrate, where the first thickness is different than the second thickness.

Abstract: Disclosed embodiments relate to instructions for fused multiply-add (FMA) operations with variable-precision inputs. In one example, a processor to execute an asymmetric FMA instruction includes fetch circuitry to fetch an FMA instruction having fields to specify an opcode, a destination, and first and second source vectors having first and second widths, respectively, decode circuitry to decode the fetched FMA instruction, and a single instruction multiple data (SIMD) execution circuit to process as many elements of the second source vector as fit into an SIMD lane width by multiplying each element by a corresponding element of the first source vector, and accumulating a resulting product with previous contents of the destination, wherein the SIMD lane width is one of 16 bits, 32 bits, and 64 bits, the first width is one of 4 bits and 8 bits, and the second width is one of 1 bit, 2 bits, and 4 bits.

Abstract: A transistor includes a body of semiconductor material, where the body has laterally opposed body sidewalls and a top surface. A gate structure contacts the top surface of the body. A source region contacts a first one of the laterally opposed body sidewalls and a drain region contacts a second one of the laterally opposed body sidewalls. A first isolation region is under the source region and has a top surface in contact with a bottom surface of the source region. A second isolation region is under the drain region and has a top surface in contact with a bottom surface of the drain region. Depending on the transistor configuration, a major portion of the inner-facing sidewalls of the first and second isolation regions contact respective sidewalls of either a subfin structure (e.g., FinFET transistor configurations) or a lower portion of a gate structure (e.g., gate-all-around transistor configuration).

Abstract: Gate-all-around integrated circuit structures having an insulator fin on an insulator substrate, and methods of fabricating gate-all-around integrated circuit structures having an insulator fin on an insulator substrate, are described. For example, an integrated circuit structure includes an insulator fin on an insulator substrate. A vertical arrangement of horizontal semiconductor nanowires is over the insulator fin. A gate stack surrounds a channel region of the vertical arrangement of horizontal semiconductor nanowires, and the gate stack is overlying the insulator fin. A pair of epitaxial source or drain structures is at first and second ends of the vertical arrangement of horizontal semiconductor nanowires and at first and second ends of the insulator fin.

Abstract: Integrated circuit structures having source or drain structures with abrupt dopant profiles are described. In an example, an integrated circuit structure includes a vertical arrangement of horizontal nanowires. A gate stack is around the vertical arrangement of horizontal nanowires. A first epitaxial source or drain structure is at a first end of the vertical arrangement of horizontal nanowires. A second epitaxial source or drain structure is at a second end of the vertical arrangement of horizontal nanowires. The first and second epitaxial source or drain structures include silicon, phosphorous and arsenic, with an atomic concentration of phosphorous substantially the same as an atomic concentration of arsenic.

Abstract: Methods and systems include a neural network system that includes a neural network accelerator. The neural network accelerator includes multiple processing engines coupled together to perform arithmetic operations in support of an inference performed using the deep neural network system. The neural network accelerator also includes a schedule-aware tensor data distribution circuitry or software that is configured to load tensor data into the multiple processing engines in a load phase, extract output data from the multiple processing engines in an extraction phase, reorganize the extracted output data, and store the reorganized extracted output data to memory.

Abstract: An embodiment of an electronic processing system may include an application processor, persistent storage media communicatively coupled to the application processor, and a graphics subsystem communicatively coupled to the application processor. The graphics subsystem may include a first graphics engine to process a graphics workload, and a second graphics engine to offload at least a portion of the graphics workload from the first graphics engine. The second graphics engine may include a low precision compute engine. The system may further include a wearable display housing the second graphics engine. Other embodiments are disclosed and claimed.

Abstract: Microelectronic assemblies, and related devices and methods, are disclosed herein. For example, in some embodiments, a microelectronic assembly may include a package substrate including a dielectric material having a first surface and an opposing second surface, a first material on at least a portion of the second surface, and a second material on at least a portion of the first material, wherein the second material has a different material composition than the first material.

Abstract: Microelectronic assemblies, related devices and methods, are disclosed herein. In some embodiments, a microelectronic assembly may include a package substrate, having a surface, including a first conductive pathway electrically coupled to a power source; a first microelectronic component, having an active side electrically coupled to the surface of the package substrate and an opposing back side, surrounded by an insulating material; a second microelectronic component, having an active side electrically coupled to the surface of the package substrate and an opposing back side, surrounded by the insulating material and including a through-substrate via (TSV) electrically coupled to the first conductive pathway; and a redistribution layer (RDL), on the insulating material, including a second conductive pathway electrically coupling the TSV, the second surface of the second microelectronic component, and the second surface of the first microelectronic component.

Abstract: Porous liners for through-glass vias and associated methods are disclosed. An example apparatus includes a glass layer having a through-hole. The example apparatus further includes a conductive material within the through-hole. The example apparatus also includes a porous material between at least a portion of the conductive material and at least a portion of a sidewall of the through-hole.

Abstract: A cross-domain device includes a first interface to couple to a first device and a second interface to couple to a second device, where the first device is to implement a first component in a radio access network (RAN) system in a first computing domain, and the second device is to implement a second component in the RAN system in a second computing domain. The first component is to interface within the second component in a RAN processing pipeline. The cross-domain device further comprises hardware to implement a communication channel between the first device and the second device to pass data from the first component to the second component, where the communication channel enforces isolation of the first computing domain from the second computing domain.

Abstract: The present disclosure is related to framework for automatically and efficiently finding machine learning (ML) architectures that are optimized to one or more specified performance metrics and/or hardware platforms. This framework provides ML architectures that are applicable to specified ML domains and are optimized for specified hardware platforms in significantly less time than could be done manually and in less time than existing ML model searching techniques. Furthermore, a user interface is provided that allows a user to search for different ML architectures based on modified search parameters, such as different hardware platform aspects and/or performance metrics. Other embodiments may be described and/or claimed.

Abstract: Solder materials and microelectronic devices and systems deploying the solder materials are discussed. The solder material includes a bulk material of tin and bismuth and particles interspersed in the tin and bismuth bulk material. The particles are a metal other than tin and bismuth, and an intermetallic compound is formed around the particles. The intermetallic compound includes the metal of the particles and tin or bismuth. The solder materials are deployed as interconnect structures to interconnect components, such as electrically coupling an integrated circuit package to a motherboard.

Abstract: Systems, apparatus, articles of manufacture, and methods are disclosed to partition a boot drive for two or more processor circuits. An example apparatus includes at least one first processor circuit to determine at least one first parameter for a first namespace and at least one second parameter for a second namespace to be configured for a non-volatile memory (NVM) boot drive, cause a first controller of the NVM boot drive to create the first namespace based on the at least one first parameter, and cause the first controller to create the second namespace based on the at least one second parameter. Also, the example at least one first processor circuit is to attach the first namespace to the first controller of the NVM boot drive, attach the second namespace to a second controller of the NVM boot drive, and attach the second controller to a bootloader of a second processor circuit.

Abstract: A cold plate comprises a plurality of fins. The individual fins have an opening, and the openings collectively define a first channel through the plurality of fins. During operation of an integrated circuit component attached to the cold plate, coolant is pumped through the cold plate. The coolant flows in a first direction through the first channel and then in a second through second channels located between the fins. The first direction is substantially orthogonal to the second direction. The first channel can comprise a tube that has openings that direct coolant to flow into the second channels. The first channel is located close to the base plate of the cold plate so that there is a high degree of heat transfer between an integrated circuit component attached to the cold plate and coolant flowing through the cold plate.

Abstract: Systems, apparatus, articles of manufacture, and methods are disclosed to implement personalized skin tone adaptation for images and video. An example apparatus disclosed herein obtains an initial skin tone group distribution for an identified user depicted in an input image. The example apparatus also determines, based on the input image, a plurality of skin tone measurements associated respectfully with a plurality of skin tone groups corresponding to the initial skin tone group distribution. The example apparatus further outputs a revised skin tone group distribution based on the skin tone measurements, the initial skin tone group distribution, and a transition model.

Abstract: Examples to determine a dynamic batch size of a layer are disclosed herein. An example apparatus to determine a dynamic batch size of a layer includes a layer operations controller to determine a layer ratio between a number of operations of a layer and weights of the layer, a comparator to compare the layer ratio to a number of operations per unit of memory size performed by a computation engine, and a batch size determination controller to, when the layer ratio is less than the number of operations per unit of memory size, determine the dynamic batch size of the layer.