Abstract: Various embodiments improve the operation of computers by providing methods of transmitting data with low latency and high bandwidth. Data may be transmitted in a packet composed of data flits, the data flits having at least two different formats configured to implement different communication protocols. In some embodiments, a given flit may be transmitted using two different modulation methods, with a first part of the flit transmitted using a first modulation method, such as a binary method, and a second part of the flit using a higher-order modulation method.

Abstract: Systems and methods are provided for managing data transmissions in integrated circuits using a handshake, credit path, and data path. For example, the handshake between transmission and receiving sides can enable information to be passed over a separate path from the data path. Based on a calculated round trip delay, the system can identify the number of top level flops or registers needed in the credit path, which can be the same as the number of top level flops or registers in the data path. The receiving side can calculate the required number of credits per requested virtual channel based on the available queuing resources of the receiving side and the round trip delay of the data path, in order to ensure full bandwidth data streaming on the channel.

Abstract: A network interface controller (NIC) capable of efficient load balancing among the hardware engines is provided. The NIC can be equipped with a plurality of ordering control units (OCUs), a queue, a selection logic block, and an allocation logic block. The selection logic block can determine, from the plurality of OCUs, an OCU for a command from the queue, which can store one or more commands. The allocation logic block can then determine a selection setting for the OCU, select an egress queue for the command based on the selection setting, and send the command to the egress queue.

Abstract: An example fixing unit includes a flexible fixing belt, a backup member located outside the fixing belt to form a fixing nip, and a plate heater in contact with the fixing belt to heat the fixing belt in the fixing nip, wherein the fixing belt includes a first abrasion prevention layer disposed on a surface facing the plate heater and including polyetheretherketone (PEEK), and the plate heater includes a second abrasion prevention layer in contact with the first abrasion prevention layer and including glass.

Abstract: A process includes coupling a push-pull driver of a first bus device to a plurality of communication lines that are associated with a first bus to allow the first bus device to access the first bus using push-pull signaling. The process includes sharing a set of communication lines with the second bus. The sharing includes coupling an open drain driver of a second bus device to the set of communication lines to allow the second bus device to access the second bus using open drain signaling. The sharing includes using the set of communication lines in first time periods in which the first bus device accesses the first bus and using the set of communication lines in second time periods other than the first time periods in which the second bus device accesses the second bus. The sharing includes isolating a push-pull driver of the first bus device from the set of communication lines responsive to the second time periods.

Abstract: An air baffle assembly for air cooling in computer systems is described. The air baffle assembly can include a baffle body removably installed on a printed circuit assembly deployed on a chassis. The baffle body can cover memory components. The air baffle assembly can also include an air dam removably installed on the baffle body. The baffle body has a slot positioned on a side of the baffle body, which can be substantially parallel to a wall of the chassis. The baffle body can further include alignment rails positioned in the slot to guide installation of the air dam into the slot. The installed air dam can reduce a bypass region, thereby reducing airflow through the bypass region and increasing airflow through the memory components.

Abstract: During operation, a layer-3 (L3) router can receive a trace query to trace multicast paths associated with a multicast group. In response to detecting a layer-2 (L2) switch coupled to the L3 router, the L3 router can construct a request packet and send the request packet to the coupled L2 switch. The request packet comprises a first data block indicating that the request packet is an L2 multicast trace request, thereby causing the L2 switch to propagate the request packet to downstream L2 switches until the request packet reaches a last-hop L2 switch. The L3 router can receive, from the last-hop L2 switch, a reply packet responsive to the L2 multicast trace request, the reply packet comprising one or more data blocks, with each data block comprising information about an L2 switch or client device present on a traversed multicast path for the multicast group.

Abstract: A fluid ejection controller interface includes input logic to receive control data packets and a first clock signal, each control data packet including a set of primitive data bits and a set of random bits, wherein the input logic identifies the random bits in the received control data packets to facilitate the creation of modified control data packets. The fluid ejection controller interface includes a clock signal generator to generate a second clock signal that is different than the first clock signal, and output logic to receive the modified control data packets, and output the modified control data packets to a fluid ejection controller of a fluid ejection device based on the second clock signal.

Abstract: An example fluid ejection device comprises a plurality of distinct fluid channels. Each fluid channel comprises a distinct fluid inlet to the ejection device. A subset of the plurality of distinct fluid channels comprises fluid recirculating fluid channels, and the remaining fluid channels comprising non-recirculating fluid channels.

Abstract: Systems, methods, and computer readable mediums for persistent TWT sessions include establishing, by an access point, a plurality of sessions with a plurality of stations, wherein one or more of the plurality of sessions comprise a target wake time (TWT), determining, by the access point, that a downtime event will occur before a TWT of a particular station of the plurality of stations, in response to the downtime event occurring during the TWT of the particular station, sending, by the access point, information associated with the plurality of sessions to a secondary device, after the downtime event, retrieving, by the access point, the information associated with the plurality of sessions from the secondary device, and reestablishing, by the access point using the information, the plurality of sessions.

Abstract: In an example, a network switch is to receive a loop detect packet from an access netwssork connected to a Data center network (DCN). The DCN includes a VXLAN overlay and the network switch is configured as a VTEP. The network switch compares the VNI of a source VTEP from which the loop detect packet originates with a locally configured VNI. In response to a match, it is determined that the network switch is configured as a peer VTEP. Import RT in the loop detect packet is compared with an export RT of the peer VTEP and the export RT in the loop detect packet is compared with an import RT of the peer VTEP. Based on the comparison, it is determined whether a VXLAN tunnel is configured between the peer and the source VTEPs. In response to the VXLAN tunnel being configured, the switch may determine that a network loop is present.

Abstract: In an example, a non-transitory computer readable storage medium may include instructions that when executed cause a computing device to generate a virtual device representation of an image forming apparatus on a cloud. Further, instructions cause the computing device to update the virtual device representation on the cloud with a configuration state of the image forming apparatus. Furthermore, instructions cause the computing device to analyze the updated virtual device representation to determine whether the image forming apparatus is compliant with a stored policy. Further, instructions cause the computing device to initiate, via the virtual device representation, a remediation action to bring the image forming apparatus to compliance based on the determination.

Abstract: A network interface controller (NIC) capable of efficient memory access is provided. The NIC can be equipped with an operation logic block, a signaling logic block, and a tracking logic block. The operation logic block can maintain an operation group associated with packets requesting an operation on a memory segment of a host device of the NIC. The signaling logic block can determine whether a packet associated with the operation group has arrived at or departed from the NIC. Furthermore, the tracking logic block can determine that a request for releasing the memory segment has been issued. The tracking logic block can then determine whether at least one packet associated with the operation group is under processing in the NIC. If no packet associated with the operation group is under processing in the NIC, tracking logic block can notify the host device that the memory segment can be released.

Abstract: One aspect of the instant application describes a system that includes a plurality of stacked mezzanine boards communicatively coupled to a motherboard and a metal enclosure enclosing the motherboard and mezzanine boards. A respective mezzanine board can include a number of solder pads, and the metal enclosure can include a plurality of metal strips, a respective metal strip to make contact with a solder pad of a corresponding mezzanine board. The system can further include a logic module positioned on the respective mezzanine board to determine a location of the respective mezzanine board based on a contact pattern between the metal strips and solder pads of the respective mezzanine board.

Abstract: A system and a method for training non-parametric Machine Learning (ML) model instances in a collaborative manner is disclosed. A non-parametric ML model instance is trained at each of a plurality of data processing nodes to obtain a plurality of non-parametric ML model instances. Each non-parametric ML model instance developed at each data processing node is shared with each of remaining data processing nodes of the plurality of data processing nodes. Each non-parametric ML model instance is processed through a trainable parametric combinator to generate a composite model at each of the plurality of data processing nodes. The composite model is trained at each of the plurality of data processing nodes, over the respective local dataset, using Swarm learning to obtain trained composite models.

Abstract: A print apparatus is disclosed. The print apparatus includes a photoconductive surface to receive a latent image representative of an image to be printed onto a printable substrate; a charging component to apply a voltage is to be applied to the photoconductive surface as the charging component moves relative to the photoconductive surface; and processing circuitry to receive data indicative of a measurement of a first current resulting from the voltage applied by the charging component; determine, responsive to detecting an increase in the measured first current relative to a reference current, the increase being greater than a first defined threshold current, that there exists a point defect under affecting the latent image; and generate instruction data responsive to determining that there exists a point defect under affecting the latent image. A method and a machine-readable medium are also disclosed.

Abstract: A print apparatus is disclosed. In an example, the print apparatus comprises: a photoconductive surface to receive a latent image representative of an image to be printed onto a printable substrate; a plurality of print components, each print component having a surface movable relative to the photoconductive surface, wherein a current or voltage is to be applied between the print component surface and the photoconductive surface; and processing circuitry to: measure a current or voltage between each print component surface and the photoconductive surface; responsive to detecting a deviation in the measured current or voltage from a reference current or voltage in respect of any of the plurality of print components, determine that there exists a defect associated with the photoconductive surface; and determine, based on the amount of deviation of the measured current or voltage from the reference current or voltage, an indication of the size of the defect.

Abstract: Systems and methods disclosed herein may include a parallel-processing graph-database solution for protein-sequence analytics to determine a viable therapeutic for a given condition, and may include: determining a protein sequence for the given condition; using sequence database to compare a query sequence of the sequence of the given condition with sequences of other known proteins in the sequence database using the sequence database to determine a similarity of the query sequence with sequences of the other known proteins in the sequence database based on the comparison; and querying a graph database based on the similarity of sequences to identify potential therapeutics that could be have an inhibitory effect on the given condition.

Abstract: End-to-end topology stitching and representation is described. An example includes instructions for receiving, at a server, a set of configuration data for an infrastructure stack, the set of configuration data including configuration data for each of a plurality of domains of the infrastructure stack; parsing the received set of configuration data; stitching together an end-to-end topology for the plurality of domains of the infrastructure stack based at least in part on the parsed set of configuration data; and generating a representation of the end-to-end topology of the infrastructure stack.

Abstract: Examples of binder agents for a three-dimensional (3D) printing process are disclosed. In an example, the binder agent includes copper nanoparticles and a liquid vehicle. In this example, the liquid vehicle includes an antioxidant, polyethylene glycol hexadecyl ether, and a balance of water. Another example of the binder agent includes stainless steel nanoparticles and a liquid vehicle. In this example, the liquid vehicle includes polyethylene glycol hexadecyl ether, and a balance of water. Still another example of the binder agent includes nickel nanoparticles and a liquid vehicle. The liquid vehicle includes an antioxidant; a symmetric triblock copolymer including poly(ethylene oxide) and poly(propylene oxide), and a balance of water.

Abstract: Implementations of the present disclosure relate to setting a system time of an access point (AP) for server certificate validation. A method comprises obtaining a default time as a system time of the AP after an AP boots up. The method also comprises obtaining a memory time from a flash memory of the AP. The method also comprises updating the system time with the memory time obtained from the flash memory. The method also comprises validating a server certificate received from an authentication server based on the system time. The system time is synchronized with a network time if the server certificate is successfully validated based on the system time. The synchronized system time is then written into the flash memory. In this way, the authentication can be performed based on a reasonable system time even if the AP reboots.

Abstract: Examples of electronic devices are described herein. In some examples, an electronic device includes a memory and a controller. In some examples, the controller is to enable a basic input/output system (BIOS) safe mode in response to an authentication procedure. In some examples, the controller is to provide access to back-up variable service region in the memory storing default BIOS data. In some examples, the controller is to start a modified BIOS based on the default BIOS data in response to enablement of the BIOS safe mode.

Abstract: A non-transitory machine-readable storage medium stores instructions that, when executed by the machine, cause the machine to provide a firewall interface between a plurality of registers of a controller and a host interface of the controller. Providing the firewall interface includes programming the firewall interface with a plurality of firewall rules. The registers are to control functions that are performed by the controller; and the plurality of firewall rules control whether requests to access the plurality of registers are denied, allowed or modified based on features of the request.

Abstract: A method and apparatus are provided for facilitating a datatype engine (DTE) to support high performance computing. A network interface card (NIC) receives, via a message passing interface, a command to read data from a host memory. The NIC determines that the command indicates a first datatype descriptor stored in the NIC. The NIC forms, based on the command, a packet which indicates a base address and a length associated with the data to be read from the host memory and passes the packet to the DTE. The DTE generates a plurality of read requests comprising offsets from the base address and corresponding lengths based on the first datatype descriptor. The DTE passes the plurality of read requests to a direct memory access module, thereby allowing the NIC to access the host memory while eliminating copies of the data on the host during transfer of the command across a network.

Abstract: In some examples, a system receives, as part of a service to update a data store, a collection of operations that are to be performed with respect to the data store, wherein the collection of operations comprises a plurality of different types of operations for respective data blocks of the data store. Multiple operations of a respective type of the different types of operations are batched to produce a batch of operations of the respective type, and the batch of operations of the respective type is flushed to the data store. The system progressively updates a tracking data structure that tracks batching and flushing of operations of the collection of operations during the service, where the tracking data structure includes metrics representing flushes of batches of operations of the different types. After an interruption of the service, the system resumes the service using a checkpointed version of a resume portion of the tracking data structure.

Abstract: Systems are methods are provided for implementing load balancing of IPsec security associations (SAs) SAs between different heterogeneous cores, for example in a multiple controller systems. The balancing of IPsec SAs are performed using software implementations of IPsec offloading and/or hardware implementations of IPsec offloading. By balancing the processing functions related to orchestration of IPsec SAs between hardware (e.g., IPsec hardware accelerator) and software (e.g., IPsec software accelerator), the methods realize increased optimization, improved user experience, and enhanced product performance in networks using IPsec. For example, an IPsec session is initiated, and it is determined whether a resource limitation for a hardware-based IPsec accelerator has been reached. If the limitation for the hardware-based IPsec accelerator has been reached, processing for the IPsec SA for the initiated IPsec session is offloaded to a software-based IPsec accelerator.

Abstract: A device and corresponding method are provided to provide accurate estimates of how long it will take to install updates to compute nodes in a large-scale computer deployment. a duration prediction model is trained using historical data from previous updates to compute nodes. The features selected to train the duration prediction model are update features including update component type, update component size, update component duration and compute node features including operating system, BMC type/version, UEFI type/version, and generation for each of the compute nodes updated. The historical data for the features is accessed from a metadata store.

Abstract: Example implementations relate to deduplication operations in a storage system. An example includes generating a housekeeping work map to delete a backup item stored in a deduplication storage system; selecting a first work entry of the housekeeping work map, where the first work entry identifies a first container index and a first manifest; in response to a selection of the first work entry, loading the first container index into the memory, the first container index comprising a back-reference data structure; identifying, in the back-reference data structure, a back-reference entry indexed to the first manifest; determining, using the back-reference entry indexed to the first manifest, a first set of data units included in the first manifest and that are indexed in the first container index; and decrementing, in the first container index, a set of reference counts for the determined first set of data units.

Abstract: In implementations of the present disclosure, there is provided a method for access point (AP) validation. The method comprises detecting a trigger event for validating a network including a plurality of APs, and selecting a target AP from the plurality of APs based on a neighbor table of an AP associated with the detected trigger event. The method further comprises determining a validation case corresponding to the detected trigger event from a knowledge base. The method further comprises sending the validation case to the target AP, and, receiving a result of performance of the validation case from the target AP in response to the validation case being performed. Implementations of the present disclosure can validate whether the network change takes effect in the network automatically, and can improve the validation efficiency of the network changes.

Abstract: Systems and methods are provided for updating resource allocation in a distributed network. For example, the method may comprise allocating a plurality of resource containers in a distributed network in accordance with a first distributed resource configuration. Upon determining that a processing workload value exceeds a stabilization threshold of the distributed network, determining a resource efficiency value of the plurality of resource containers in the distributed network. When a resource efficiency value is greater than or equal to the threshold resource efficiency value, the method may generate a second distributed resource configuration that includes a resource upscaling process, or when the resource efficiency value is less than the threshold resource efficiency value, the method may generate the second distributed resource configuration that includes a resource outscaling process. The resource allocation may transmit the second to update the resource allocation.

Abstract: Example implementations relate to microphone directional beamforming adjustments. In some examples, a computing device can include a microphone, a wireless receiver, and a processor, where the processor is to receive a wireless signal from a peripheral device, determine a location of the peripheral device based on a signal strength between the peripheral device and the wireless receiver, and cause, based on the location of the peripheral device, the microphone to adjust a beamforming direction of the microphone.

Abstract: A device includes at least one well to receive a polymerase chain reaction (PCR) mixture and including a bottom. The bottom includes an electrically resistive sheet which has a relative magnetic permeability no greater than about 1.01, wherein the resistive sheet is to receive a signal from a signal source to cause the resistive sheet to generate heat to form a pulse-controlled amplification, thermal cycling zone in close thermal proximity to the bottom.

Abstract: Examples of methods are described herein. In some examples, a method includes determining signal variation data of a fluorescence signal measured from an amplification procedure of a nucleic acid sample. In some examples, the method includes detecting, using a machine learning model, a target nucleic acid strand in the nucleic acid sample based on the signal variation data.

Abstract: In an example in accordance with the present disclosure, a fiber molding screen is described. The fiber molding screen includes a first region and a second region. The fiber molding screen includes a first set of pores formed within the first region. Each pore of the first set of pores has a first pore trajectory angle between a longitudinal axis of the pore and the first region surface. The fiber molding screen also includes a second set of pores formed within the second region. Each pore of the second set of pores has a second pore trajectory angle between a longitudinal axis of the pore and the second region surface. The second pore trajectory angle is different than the first pore trajectory angle.

Abstract: Examples described herein relate to a security management system to secure a container ecosystem. In some examples, the security management system may protect one or more entities such as container management applications, container images, containers, and/or executable applications within the containers. The security management system may make use of digital cryptography to generate digital signatures corresponding to one or more of these entities and verify them during the execution so that any compromised entities can be blocked from execution and the container ecosystem may be safeguarded from any malicious network attacks.

Abstract: An example three-dimensional (3D) printer may include a camera to capture a low-resolution thermal image of a build material bed. The 3D printer may include an interpolation engine to generate an interpolated thermal image based on the low-resolution thermal image. The 3D printer may also include a correction engine to enhance fine details of the interpolated thermal image without distorting thermal values from portions of the interpolated thermal image without fine details to produce an enhanced thermal image.

Abstract: A system for policy management in a switch is provided. During operation, the system can generate, from a first policy defined for the switch, a second policy. The first policy can indicate whether a type of traffic is allowed from a source role to a destination role via an overlay tunnel. The second policy can indicate a plurality of destination roles that are allowed to receive multi-destination packets of the type of traffic from the source role via the overlay tunnel. Upon identifying a host associated with a role at a port of the switch, the system can determine whether the role belongs to the plurality of destination roles based on the second policy. If the role belongs to the plurality of allowed destination roles, the system can allow the port to forward a multi-destination packet, which is received via the overlay tunnel and associated with the type of traffic.

Abstract: In an example implementation consistent with the features disclosed herein, publisher-subscriber message mapping is performed. A message is received from a publisher, with the message including a first topic. The first topic includes a gateway identifier of a gateway for the publisher. The gateway identifier is then mapped to a user identifier. The first topic of the message is modified to a second topic by replacing the gateway identifier with the user identifier. The message is then sent to a subscriber that is subscribed to the second topic.

Abstract: A network infrastructure management console configures one or more network switch stacks each comprising a plurality of switches. A monitoring component monitors a current conductor switch of the stack. A user interface (UI) backend component comprises a cache memory and receives a user request to configure the stack. The UI backend component receives from the monitoring component notification of the current conductor and stores in a cache memory segment associated with the current conductor the requested configuration changes. If the current conductor switch of the stack has changed due to a failover event, the configuration changes stored in the cache memory segment associated with the previous current conductor are written to a cache memory segment associated with the new current conductor. A configuration push component receives the configuration changes and transmits the configuration changes to the network switch stack.

Abstract: Systems are methods are provided for implementing an intelligent optical transceiver. The intelligent optical transceiver implements dynamic health monitoring and “time to fail” prediction functions to predict a failure of a component before it malfunctions during use. By employing the intelligent optical transceiver, a network can prevent failures in its optical connectivity that can degrade the network performance, such as experiencing outages and data unavailability. For example, the intelligent optical transceiver includes a module health monitor, which monitors, in real-time, health parameters of optical communication components in the intelligent optical transceiver. Also, the intelligent optical transceiver includes a time to fail predictor which predicts a time to fail for the optical communication components of the optical transceiver based on a result of a defined regression function.

Abstract: Example implementations relate to optimizing operation of microservice clusters comprising multiple nodes, each executing a common self-sufficient microservice instance. The method includes, at a first node having a database instance comprising a plurality of rows and a plurality of columns, calculating a distinct hash per row to create a hash list, each hash identifying data contained in the respective row; publishing a distinct hash and/or the hash list to one or more of the plurality of nodes, each node having respective database instances and respective hash lists; at a second node, comparing the distinct hash and/the hash list published by the first node to the hash list of the second node to identify any missing rows of data; and, in response to identifying, based on the comparison, a missing row(s) in the second node's database instance, updating the second node's database instance to include the missing row(s) of data.

Abstract: A system that can intelligently help a user judge the resolution requirement of the image content. In an example, the display panel can show in different virtual resolutions or the system can automatically judge whether the image content requires high or low resolution using a similarity judgement. The system can then inform the display panel what virtual resolution should be shown or what the optimal resolution is using the similarity judgment and send the image content to the display panel with that resolution.

Abstract: Systems and methods for providing remote network security using a network embeddings model are provided. A method consistent with the present disclosure includes retrieving a corpus of network activity data associated with a first network. The network activity data may be generated from users within the first network submitting network requests for network assets to service the network requests. The method also includes creating a crafted encoded corpus by selecting a subset of the corpus of network activity data and creating a network embeddings model based on the crafted encoded corpus. Lastly, the method includes generating an alert in an event that the network security system identifies an anomaly associated with the crafted encoded corpus of network activity data.

Abstract: An example device is described for facilitating polygon localization. In various aspects, the device can comprise a processor. In various instances, the device can comprise a non-transitory machine-readable memory that can store machine-readable instructions. In various cases, the processor can execute the machine-readable instructions, which can cause the processor to localize a polygon depicted in an image, based on execution of a deep learning pipeline. In various aspects, the deep learning pipeline can comprise a circular-softmax block.

Abstract: The present disclosure pertains to systems and methods for mapping an image to a three-dimensional pattern of relief features on a surface of an object for 3D printing. In some examples, a system may determine a pattern of relief features, including debossed relief features and/or embossed relief features, to create a shading pattern on the surface of 3D printed object that approximates the continuous halftones of an image.

Abstract: A system and a method of preventing generation of duplicate network routes in a Software Defined Wide Area Network (SD-WAN) are described. A network route management service receives network route information from a network device connected to a subnetwork present in the SD-WAN. The network route information is compared with a master network route information for identifying presence of one or more duplicate network routes in the network route information. Upon identification, the duplicate network routes are removed from the network route information, and the network route information excluding the one or more duplicate network routes is transmitted to the network device.

Abstract: Power-over-Ethernet (PoE) powered devices (PD) may be coupled to a PoE power sourcing equipment (PSE). The PDs may send to the PSE a link-layer protocol communication that comprises an alternative-power field indicating whether the sending PD has an alternative power source (e.g., a battery, a local power supply). The PSE may listen for and receive the communications and read the alternative-power fields thereof. The PSE may set respective power priorities for the PDs based at least in part on whether the PDs have respective alternative power sources, as indicated by the respective alternative-power fields of their communications. The PSE may reduce the power priority of those PDs that have alternative power sources, relative to the priority the PD would otherwise have been given.

Abstract: In a three-dimensional printing method example, a liquid functional agent is selectively applied. The liquid functional agent includes an alloying agent. A metallic build material is applied. The liquid functional agent is selectively applied before the metallic build material, after the metallic build material, or both before and after the metallic build material. The liquid functional agent patterns the metallic build material to form a composite layer. At least some of the metallic build material is exposed to energy to melt the at least some of the metallic build material to form a layer. Upon contact or after energy exposure, the alloying agent and the build material alter a composition of the composite layer.

Abstract: A system determines a first set of policies, wherein at least one policy entry for a destination role comprises a source role, a traffic attribute, and an action to be taken for the packet. The system represents the policies as a matrix, wherein a first entry in the matrix indicates the source and destination role, the traffic attribute, and the action of the at least one policy entry. The system replaces, in the first entry, the action with the destination role if the action indicates to allow the packet, and with a null value if the action indicates to deny the packet, to obtain a first data structure with entries indicating, for a respective source role, traffic attributes and corresponding sets of allowed destination roles. The system resolves an overlapping pair comprising a first and a second traffic attribute to obtain a second set of synthesized policies.