Abstract: Techniques and systems disclosed herein may relate to a server processing work requests of a work requester. For example, the work requester may generate a plurality of work requests comprising functions of an application to be executed by a dependency-aware (DA) server, determine processing dependencies associated with the work requests, communicate, to the DA server, the processing dependencies, and enqueue the work requests on a dependent queue and a conditioned dependent queue based on processing dependencies associated with the DA server. The DA server may then select one of the work requests from the dependent queue or the conditioned dependent queue for processing and processing the selected work request.

Abstract: A thermoelectric cooler (TEC) is positioned to move heat away from a hot spot on a semiconductor chip and toward a dielectric substrate. This approach to thermal management is particularly effective when used in conjunction with a buried rail and back side power delivery. The TEC may be in a layer that contains solder connections be between two device layers an IC package. Alternatively, the TEC may be in a metal interconnect structure over the semiconductor substrate such as in a passivation stack at the top of the metal interconnect structure. TECs at either of these locations may be formed by wafer-level processing.

Abstract: Epitaxial source/drain structures for enhancing performance of multigate devices, such as fin-like field-effect transistors (FETs) or gate-all-around (GAA) FETs, and methods of fabricating the epitaxial source/drain structures, are disclosed herein. An exemplary device includes a dielectric substrate. The device further includes a channel layer, a gate disposed over the channel layer, and an epitaxial source/drain structure disposed adjacent to the channel layer. The channel layer, the gate, and the epitaxial source/drain structure are disposed over the dielectric substrate. The epitaxial source/drain structure includes an inner portion having a first dopant concentration and an outer portion having a second dopant concentration that is less than the first dopant concentration. The inner portion physically contacts the dielectric substrate, and the outer portion is disposed between the inner portion and the channel layer. In some embodiments, the outer portion physically contacts the dielectric substrate.

Abstract: One aspect of the present disclosure pertains to a semiconductor device. The semiconductor device includes a semiconductor substrate and a transistor formed over the semiconductor substrate. The transistor includes a first source/drain (S/D) feature, a second S/D feature, a channel region interposed between the first and second S/D features, and a gate stack engaging the channel region. The semiconductor device includes a first S/D contact landing on a top surface of the first S/D feature, a second S/D contact landing on a top surface of the second S/D feature, and a dielectric plug penetrating through the semiconductor substrate and landing on a bottom surface of the first S/D feature. The dielectric plug spans a width equal to or smaller than a width of the first S/D feature.

Abstract: A semiconductor structure can include a high voltage region, a first moat trench isolation structure electrically insulating the high voltage region from low voltage regions of the semiconductor structure, and a second moat trench isolation structure electrically insulating the high voltage region from the low voltage regions of the semiconductor structure. The first moat trench isolation structure can include dielectric sidewall spacers and a conductive fill material portion located between the dielectric sidewall spacers. The second moat trench isolation structure can include only at least one dielectric material, and can include a dielectric moat trench fill structure having a same material composition as the dielectric sidewall spacers and having a lateral thickness that is greater than a lateral thickness of the dielectric sidewall spacers and is less than twice the lateral thickness of the dielectric sidewall spacers.

Abstract: An interconnect fabrication method is disclosed herein that utilizes a disposable etch stop hard mask over a gate structure during source/drain contact formation and replaces the disposable etch stop hard mask with a dielectric feature (in some embodiments, dielectric layers having a lower dielectric constant than a dielectric constant of dielectric layers of the disposable etch stop hard mask) before gate contact formation. An exemplary device includes a contact etch stop layer (CESL) having a first sidewall CESL portion and a second sidewall CESL portion separated by a spacing and a dielectric feature disposed over a gate structure, where the dielectric feature and the gate structure fill the spacing between the first sidewall CESL portion and the second sidewall CESL portion. The dielectric feature includes a bulk dielectric over a dielectric liner. The dielectric liner separates the bulk dielectric from the gate structure and the CESL.

Abstract: A semiconductor structure can include a first substrate having a frontside and a backside opposite the frontside. The semiconductor structure can include devices on the frontside. The semiconductor structure can include first interconnect structures on the frontside and coupled to the devices. The semiconductor structure can include a heat distribution layer on the frontside and electrically isolated from the first interconnect structures, where the heat distribution layer includes a thermally conductive material. The semiconductor structure can include a second substrate coupled to the first substrate on the frontside. The semiconductor structure can include second interconnect structures on the backside and coupled to the devices.

Abstract: A circuit module with improved timing control, may comprise a functional circuit, a control circuit, a main auxiliary circuit and an additional auxiliary circuit. The control circuit may control operation timing of the functional circuit according to response characteristics of a first node. When enabled, the main auxiliary circuit may provide main conduction path(s) between the first node and a base node. Respectively when enabled and disabled, the additional auxiliary circuit may provide and not provide additional conduction path(s) between the first node and the base node. When the control circuit controls the operation timing of the functional circuit, the main auxiliary circuit may be enabled, and the additional auxiliary circuit may be disabled or enabled according to whether a mode signal is of a first mode level or a second level.

Abstract: A camera structure, including a lens holder, a lens frame and a plurality of balls. The lens holder has a holder body, one end of which has a first rolling groove. The first groove wall part and the second groove wall part are disposed on two sides of the first rolling groove, and the groove bottom is disposed between the first groove wall part and the second groove wall part. The lens frame is mounted on an outer side of the holder body. The plurality of balls are located inside the first rolling groove, wherein the first groove wall part and the second groove wall part support the plurality of balls, there is a gap between each of the plurality of balls and the groove bottom, and the plurality of balls lay between the lens holder and the lens frame.

Abstract: A semiconductor die and the method of forming the same are provided. The semiconductor die includes a first interconnect structure, a second interconnect structure including a conductive feature, and a device layer between the first interconnect structure and the second interconnect structure. The device layer includes a semiconductor fin, a first gate structure on the semiconductor fin, a source/drain region adjacent the first gate structure, and a shared contact extending through the semiconductor fin to be electrically connected to the source/drain region and the first gate structure. The conductive feature contacts the shared contact.

Abstract: A MRAM layout structure with multiple unit cells, including a first word line, a second word line and a third word line extending through active areas, wherein two ends of a first MTJ are connected respectively to a second active area and one end of a second MTJ, and two ends of a third MTJ are connected respectively to a third active area and one end of a fourth MTJ, and a first bit line and a second bit line connected respectively to the other end of the second MTJ and the other end of the fourth MTJ.

Abstract: A drive tray assembly is disclosed for placing drives (e.g., hard disk drives) in a computer chassis in a rotated orientation and without the need for additional tools. A tray base holding a drive can be slid into a receiving space of the chassis in a longitudinal direction. An installation lever rotatably attached to the tray base includes a shaft that engages a corresponding receiving slot of the receiving space. After the tray base is inserted, rotation of the installation lever to a closed position causes the tray base, and thus the drive, to move in a direction perpendicular to the longitudinal direction. A release lever rotatably attached to the tray base can secure the installation lever in the closed position until the release lever is rotated, freeing the installation lever to move to the open position in which the drive tray may be removed from the chassis.

Abstract: A magnetoresistive random access memory (MRAM) device includes a first array region and a second array region on a substrate, a first magnetic tunneling junction (MTJ) on the first array region, a first top electrode on the first MTJ, a second MTJ on the second array region, and a second top electrode on the second MTJ. Preferably, the first top electrode and the second top electrode include different nitrogen to titanium (N/Ti) ratios.

Abstract: An E-paper display panel including an E-paper display layer, a first substrate, a pixel array layer, a common electrode layer, and a driving circuit is provided. The first substrate is disposed at a first side of the E-paper display layer. The pixel array substrate is disposed between the first substrate and the E-paper display layer and includes touch electrodes and driving pixels arranged in an array. Each driving pixel includes a first pixel electrode and a second pixel electrode. The touch electrodes, the first pixel electrode, and the second pixel electrode are overlapped with each other. The common electrode layer is disposed at a second side of the E-paper display layer. The first side is opposite to the second side. The driving circuit is in signal communication with the common electrode layer and the pixel array layer. The touch electrodes are individually in signal communication with the driving circuit.

Abstract: A semiconductor structure and a method of forming the same are provided. An exemplary method of forming the semiconductor structure includes receiving a workpiece including a fin structure over a front side of a substrate, recessing a source region of the fin structure to form a source opening, extending the source opening into the substrate to form a plug opening, forming a semiconductor plug in the plug opening, planarizing the substrate to expose the semiconductor plug from a back side of the substrate, performing a first wet etching process to remove a portion of the substrate, performing a pre-amorphous implantation (PAI) process to amorphize a rest portion of the substrate, performing a second wet etching process to remove the amorphized rest portion of the substrate to form a dielectric opening, depositing a dielectric layer in the dielectric opening, and replacing the semiconductor plug with a backside source contact.

Abstract: A memory module with improved timing adaptivity of sensing amplification, comprises at least one sensing amplifier, a tracking word line, a tracking bit line and a pulse-width controller. The tracking word line comprises a front node and an end node. Each said sensing amplifier is enabled/disabled when an enabling signal is activated/deactivated. The pulse-width controller is coupled to the tracking bit line, the front node and the end node. When a voltage of the tracking bit line changes to a predetermined voltage, the pulse-width controller activates the enabling signal, and causes a voltage of the front node to change. When the voltage of the front node changes, the tracking word line causes a voltage of the end node to change after a first delay time. When the voltage of the end node changes, the pulse-width controller deactivates the enabling signal after a second delay time.

Abstract: A computer implemented method comprises a server processing work requests of a work requester. The work requester can communicate to the server a processing dependency of one work request on a second work request. The server can associate the dependency with the work requests and/or a queue of work requests. The dependency include a condition to be met in association with processing the work requests, and the condition can include an action for the server to take in association with processing a work request. A computing system can comprise a work requester, a server, and a set of dependency-aware queues for processing a set of work requests. A queue and/or work requests on the queues can be associated with a processing dependency and the server can process work requests enqueued to the queues in an order based on the dependencies. A work requester/server interface can comprise a dependency framework.

Abstract: In a method an Intelligent Data Conversion (IDC) engine of a dataflow system detects a stage transition of a dataflow application executing on the dataflow system. In response, the IDC engine determines that data among stage data of the application has a first Stage Data Format (SDF). The IDC engine determines that a first processing unit of the dataflow system can process data having a second SDF and determines a data conversion to convert data among the stage data to have the second SDF. The IDC engine also determines a second processing unit, of the dataflow system to perform the data conversion and dispatches the second processing unit to perform the data conversion. The dataflow computing system can include a runtime processor and the IDC engine can interact with the runtime processor to detect the stage transition and/or dispatch the first processing unit.

Abstract: A memory module with improved timing adaptivity of sensing amplification, comprises at least one sensing amplifier, a tracking word line, a tracking bit line and a pulse-width controller. The tracking word line comprises a front node and an end node. Each said sensing amplifier is enabled/disabled when an enabling signal is activated/deactivated. The pulse-width controller is coupled to the tracking bit line, the front node and the end node. When a voltage of the tracking bit line changes to a predetermined voltage, the pulse-width controller activates the enabling signal, and causes a voltage of the front node to change. When the voltage of the front node changes, the tracking word line causes a voltage of the end node to change after a first delay time. When the voltage of the end node changes, the pulse-width controller deactivates the enabling signal after a second delay time.

Abstract: In a method an orchestrator of a computing system determines that results of Machine Learning model computations are available and dispatches a worker to perform model computations that include computing gradients of the results. The orchestrator determines that a set of gradients of the results is available and dispatches a gradient worker to compute a sum of the gradients. The orchestrator determines that a second set of gradients of the results is available and dispatches a second gradient worker to compute a sum of the second set of gradients. The orchestrator determines that the sums of the first and second gradients are available and dispatches a third gradient worker to compute synchronized gradients. The gradient workers compute the sums and synchronized gradients concurrent with training workers computing additional model computations results and/or gradients. A computer program product can include the method and a computing system can include the orchestrator.