Abstract: Techniques and mechanisms for dynamically changing a distribution of credits with which initiator agents of a network access a shared target resource of the network. In various embodiments, a target agent and multiple initiator agents are coupled to each other via a switched network, and further via a credit management bus (CMB). The target agent manages a credit-based scheme according to which the initiator agents share access to a target resource. Communications via the CMB enable the target agent to determine, during a runtime of the network, whether a given initiator agent has been allocated an excessive number of credits, or an insufficient number of credits. In another embodiments, the target agent changes the distribution of credits to the initiator agents by allocating credits via the CMB.

Abstract: Methods and apparatus for an accelerator controller hub (ACH). The ACH may be a stand-alone component or integrated on-die or on package in an accelerator such as a GPU. The ACH may include a host device link (HDL) interface, one or more Peripheral Component Interconnect Express (PCIe) interfaces, one or more high performance accelerator link (HPAL) interfaces, and a router, operatively coupled to each of the HDL interface, the one or more PCIe interfaces, and the one or more HPAL interfaces. The HDL interface is configured to be coupled to a host CPU via an HDL link and the one or more HPAL interfaces are configured to be coupled to one or more HPALs that are used to access high performance accelerator fabrics (HPAFs) such as NVlink fabrics and CCIX (Cache Coherent Interconnect for Accelerators) fabrics. Platforms including ACHs or accelerators with integrated ACHs support RDMA transfers using RDMA semantics to enable transfers between accelerator memory on initiators and targets without CPU involvement.

Abstract: A processor unit comprising a first controller to couple to a host processing unit over a first link; a second controller to couple to a second processor unit over a second link, wherein the second processor unit is to couple to the host central processing unit via a third link; and circuitry to determine whether to send a cache coherent request to the host central processing unit over the first link or over the second link via the second processing unit.

Abstract: A method and system for assessing tissue to determine a presence or absence of cancer cells. The method includes acquiring fluorescence lifetime (FLT) data from tissue and processing the FLT data to determine a FLT signal at each of a plurality of locations across the tissue. The method also includes determining FLT data at any of the plurality of locations above a threshold indicative a presence of cancer cells and generating a report indicating any of the plurality of locations above the threshold as indicative the presence of cancer cells.

Abstract: Systems and methods are provided for optimizing N10 interface messaging. Specifically, registration messages, de-registration messages, and the like can be dynamically controlled such that communication is prevented. Said messages are communicated via an N10 interface between a Unified Data Management function (UDM) and a Session Management function (SMF) in a 5G network. Additionally, subscription information, user profile information, etc., is also communicated via the N10 interface. Thus, the N10 interface can easily become overloaded or congested. Aspects herein provide mechanisms to avoid additional congestion of the N10 interface by intelligently eliminating specific messages communicated to the UDM via the N10 interface.

Abstract: Systems and methods are provided for optimizing N10 interface messaging. Specifically, registration messages, de-registration messages, and the like can be dynamically controlled such that communication is prevented. Said messages are communicated via an N10 interface between a Unified Data Management function (UDM) and a Session Management function (SMF) in a 5G network. Additionally, subscription information, user profile information, etc., is also communicated via the N10 interface. Thus, the N10 interface can easily become overloaded or congested. Aspects herein provide mechanisms to avoid additional congestion of the N10 interface by intelligently eliminating specific messages communicated to the UDM via the N10 interface.

Abstract: Systems or methods of the present disclosure may provide systems and techniques for efficiently transferring data between a host processing unit and connected devices using coherent doorbell register updates. For example, a method may include: receiving, via controller of host processing circuitry, an attempt to write data to a cacheable memory address from a processing unit of the host processing circuitry; transmitting, via the controller, an indication of the attempt to write to the cacheable memory address to a device; receiving, via the controller, an acknowledgment from the device that the cacheable memory address has been deallocated by the device; and writing, via the controller, the data to the cacheable memory address in response to the acknowledgement from the device.

Abstract: Systems or methods of the present disclosure may provide systems and techniques for sharing resources of an IC device between communications pipelines of the IC device. For example, a method may include: receiving a request from a first initiator component, the request associated with a first communication protocol; storing the request in a shared buffer; receiving a response from a first target component, the response associated with a second communication protocol; storing the response in the shared buffer; sending the request from the shared buffer to a second target component; and sending the response from the shared buffer to a second initiator component.

Abstract: Systems and methods are provided for providing communication sessions to user equipment (UE) over a telecommunications network. Specifically, communication sessions can be established for the UE despite a degraded N10 interface between a Unified Data Management function (UDM) and a Session Management function (SMF) in a 5G core network. One or more profiles comprising one or more profile parameters can be stored locally at the SMF so the one or more profile parameters can be provided to establish a communication session for a UE without receiving a response from the UDM. A locally stored profile on the SMF can be utilized when the N10 interface fails, when the UDM is experiencing congestion over a predetermined threshold, etc. Criteria that may impact the ability of the UDM to respond to queries can be identified to trigger a fetch of the profile parameters and the creation of a local profile at the SMF.

Abstract: The disclosed system includes a first gateway of a first type configured to receive wireless packets from a core network node, and a second gateway of a second type colocated with the first gateway and configured to transmit at least a portion of the wireless packets to a network. The first gateway is configured with a first internet address that identifies the second gateway and a second internet address that identifies a third gateway of the second type. The first gateway is configured to transmit a first portion of the wireless packets to the second gateway based on the first internet address and transmit a second portion of the wireless packets to the third gateway based on the second internet address. The second gateway is configured with a third internet address that identifies the second gateway.

Abstract: System and methods for efficient user plane function (UPF) selection are described. A mobile device (UE) may transmit a connection request to a 5G core network. The connection request can include an indication that the type of data to be serviced is IPv4-type traffic. The access management function (AMF) receives the request from a base station (e.g., gNodeB) and queries a network function repository function (NRF) for an identification of a session management function (SMF). The selected SMF queries a session management function datastore. If the SMF selected by the NRF is capable of IPV4 traffic, the identification of the SMF is returned to the AMF. If the SMF is not capable of IPV4 traffic, the identification of a capable SMF using the information received in the SMF status datastore is returned to the AMF through a 308 code. The mobile device is thereafter connected to the capable UPF.

Abstract: Systems, devices, and techniques described herein relate to efficient Evolved Packet System (EPS) fallback. A method may include receiving a rejection message that indicates a rejection to set up a call through a 5th Generation (5G) network by a 5G Radio Access Network (RAN). In response to receiving the rejection message, the method may include transmitting a request to establish a dedicated bearer through a 4th Generation (4G) network. A confirmation that the second dedicated bearer has been established through the 4G network may be received within a predetermined time after transmitting the request.

Abstract: The authorized IP addresses configured at components in an IMS network may become out of sync with a first pool of IP addresses provided by a core network to UEs. This may result in emergency communications being received at the IMS network from a UE using a first IP address of the first pool of IP addresses that is not recognized by the IMS network. A session management component of the core network may automatically assign a second IP address from a second pool of IP addresses based on determining the session management component has not received an update notification from the IMS network within a threshold time (e.g., 500 milliseconds, 1000 milliseconds, etc.) from transmitting a request to establish a packet data communications session to the IMS network using the first IP address. The IMS network may then receive packets from the second IP address.

Abstract: A wireless communication network handles Session Management Function (SMF) outage in a Tracking Area (TA). An external SMF that is outside the TA registers with an internal Network Repository Function (NRF) that is inside in the TA. The internal NRF detects the SMF outage for an internal SMF inside the TA. The internal NRF receives an SMF request from an internal Access and Mobility Management Function (AMF) inside the TA. In response to the SMF outage, the internal NRF transfers an SMF response to the internal AMF that indicates the external SMF. The external SMF receives network signaling from the internal AMF and responsively directs an external User Plane Function (UPF) that is outside the TA to exchange user data with internal wireless access nodes that are inside the TA.

Abstract: Systems, devices, and techniques described herein relate to interception of data traffic that is traversing a network at a relatively high throughput rate. In some examples, a request to initiate interception is received from a server. In response to receiving the request to initiate interception, an instruction to forward user plane data to the server can be transmitted to a User Plane Function (UPF). The UPF may be transferring the user plane data in a data flow. The instruction may cause the UPF to forward the user plane data to the server in multiple streams. In addition, control plane data associated with the data flow can be transmitted to the server.

Abstract: A method for selecting a user plane function when a user equipment (UE) attaches to a wireless communication network may include determining a data throughput limit associated with the UE, identifying a first user plane function having a first data throughput capacity, and identifying a second user plane function having a second data throughput capacity, wherein the first throughput capacity is higher than the second throughput capacity. The method may further include selecting the first user plane function if the data throughput limit of the UE is above a predetermined threshold, selecting the second user plane function if the data throughput limit of the UE is below the predetermined threshold, and sending a session request to the selected first or second user plane function.

Abstract: An integrated circuit includes a core region of logic circuits and a network routed outside the core region. The network includes a wide layer and a narrow layer. The wide layer comprises first routers coupled in series. The narrow layer comprises second routers coupled in series.

Abstract: Systems, devices, and techniques described herein relate to migrating a communication session from a path including a stressed user plane function (UPF) to a path including a replacement UPF. A communication session may traverse a first path including the first UPF. After establishing the communication session, the first UPF may be determined to be stressed. In response, the communication session can be proactively migrated to a second path including a second UPF. According to various implementations, the existing communication session can be maintained during the migration, thereby substantially eliminating interruptions caused by the stressed first UPF.

Abstract: Methods, media, and systems provide a more robust IP Multimedia Subsystem (IMS) deregistration process that deregisters a user equipment under circumstances that currently prevent a timely deregistration. For example, the aspects deregister a user equipment (UE) from an IMS core even when the UE disconnects from the Packet Data Network Gateway (PGW) or other EPC component without sending a SIP deregistration to the IMS core. The technology described herein automatically generates a deregistration message and communicates the message to the IMS core (e.g., a Session Termination Request Communication using Diameter Protocol between a PCRF & a P-CSCF) upon determining the session with the EPC is terminated. The UE is then deregistered from the IMS core.

Abstract: Solutions for providing a data traffic session for a user equipment (UE) over a wireless network comprises: receiving, by a session management node (e.g., a session management function, SMF), for a packet routing node (e.g., a user plane function, UPF), an indication of supported interne protocol (IP) traffic type (e.g., IPv4 or IPv6), wherein the indication of supported IP traffic type indicates support for a first IP traffic type or support for a second IP traffic type; allocating, by the session management node, to the packet routing node, a set of IP addresses based on at least the indication of supported IP traffic types, wherein allocating the set of IP addresses comprises allocating only supported IP addresses and withholding allocation of unsupported IP addresses; and providing, to the UE, using an IP address of the set of IP addresses, the data traffic session through the packet routing node.