Abstract: Methods and apparatus for implementing active interconnect link power management using an adaptive low-power link-state entry policy. The power state of an interconnect link or fabric is changed in response to applicable conditions determined by low-power link-state entry policy logic in view of runtime traffic on the interconnect link or fabric. The low-power link-state policy logic may be configured to include consideration of operating system input and Quality of Service (QoS) requirements for applications and devices employing the link or fabric, and device latency tolerance requirements.

Abstract: The present disclosure provides techniques for cache management. A data block may be received from an IO interface. After receiving the data block, the occupancy level of a cache memory may be determined. The data block may be directed to a main memory if the occupancy level exceeds a threshold. The data block may be directed to a cache memory if the occupancy level is below a threshold.

Abstract: In an embodiment, a processor includes multiple tiles, each including a core and a tile cache hierarchy. This tile cache hierarchy includes a first level cache, a mid-level cache (MLC) and a last level cache (LLC), and each of these caches is private to the tile. A controller coupled to the tiles includes a cache power control logic to receive utilization information regarding the core and the tile cache hierarchy of a tile and to cause the LLC of the tile to be independently power gated, based at least in part on this information. Other embodiments are described and claimed.

Abstract: In an embodiment, a processor includes multiple tiles, each including a core and a tile cache hierarchy. This tile cache hierarchy includes a first level cache, a mid-level cache (MLC) and a last level cache (LLC), and each of these caches is private to the tile. A controller coupled to the tiles includes a cache power control logic to receive utilization information regarding the core and the tile cache hierarchy of a tile and to cause the LLC of the tile to be independently power gated, based at least in part on this information. Other embodiments are described and claimed.

Abstract: Methods and systems may provide for determining a next active window for a platform and notifying one or more of a plurality of devices of the platform of the next active window being determined. Additionally, one or more of the plurality of devices may be notified of an onset of the next active window. In one example, a pre-warm message is issued to notify one or more of the plurality of devices of the next active window being determined.

Abstract: A method and apparatus for selectively parking routers used for routing traffic in mesh interconnects. Various router parking (RP) algorithms are disclosed, including an aggressive RP algorithm where a minimum number of routers are kept active to ensure adequate network connectivity between active nodes and/or intercommunicating nodes, leading to a maximum reduction in static power consumption, and a conservative RP algorithm that favors network latency considerations over static power consumption while also reducing power. An adaptive RP algorithm is also disclosed that implements aspects of the aggressive and conservative RP algorithms to balance power consumption and latency considerations in response to ongoing node utilization and associated traffic. The techniques may be implemented in internal network structures, such as for single chip computers, as well as external network structures, such as computing clusters and massively parallel computer architectures.

Abstract: Method and apparatus for reliable multicast communication over wireless network are provided. According to embodiments of the invention, the method includes determining a priority category for a multicast communication to be transmitted. The method includes designating, for the multicast communication, one of the multicast communication recipients as a leader based on the priority category and multicast diagnostics information received from the multicast communication recipients. The leader is assigned to transmit to the multicast communication source an acknowledgment frame indicating receipt of a multicast communication frame received from the source.

Abstract: Methods and apparatus for provision of adaptive packet deflection to achieve fair, low-cost, and/or energy-efficient Quality of Service (QoS) in Network-on-Chip (NoC) devices are described. In some embodiments, it is determined whether a target port of a packet has reached a threshold utilization value and the packet is routed to an alternate port in response to a deflection probability value that is to be determined based on a utilization value of the target port and a priority level value of the packet. Other embodiments are also claimed and/or disclosed.

Abstract: Embodiments of methods, systems, and storage medium associated with are disclosed herein. In one instance, the method may include: first determining whether the computing device is connected to a network, based on a result of the first determining, monitoring data traffic between the computing device and the network, wherein the data traffic is associated with at least one application residing on the computing device, based on the monitoring, second determining whether the at least one application has been updated, and initiating a transition of the computing device to a sleep mode upon a result of the second determining that indicates that the at least one application has been updated. Other embodiments may be described and/or claimed.

Abstract: Devices and methods for optimizing semi-active workloads are described herein. A network interface device may be configured to offload data packet acknowledgment responsibilities of a host platform by transmitting, to the sender of the packets, acknowledgements of packets received throughout a time duration. Upon completion of the time duration, the network interface device may trigger the host platform to perform batch processing of the data packets received during the time duration.

Abstract: Embodiments of an apparatus, system and method are described for input/output (I/O) device assisted platform power management. An apparatus may comprise, for example, power management logic operative to receive idle duration information from one or more input/output (I/O) devices and to modify a power state for one or more components based on the idle information. Other embodiments are described and claimed.

Abstract: Systems and methods may provide for aggregating a first idle duration from a first device associated with a platform and a second idle duration from a second device associated with the platform. Additionally, an idle state may be selected for the platform based at least in part on the first idle duration and the second idle duration. In one example, the idle durations are classified as deterministic, estimated or statistical.

Abstract: The present invention may provide a computer system including a plurality of tiles divided into multiple virtual domains. Each tile may include a router to communicate with others of said tiles, a private cache to store data, and a spill table to record pointers for data evicted from the private cache to a remote host, wherein the remote host and the respective tile are provided in the same virtual domain. The spill tables may allow for faster retrieval of previously evicted data because the home registry does not need to be referenced if requested data is listed in the spill table. Therefore, embodiments of the present invention may provide a distance-aware cache collaboration architecture without incurring extraneous overhead expenses.

Abstract: A mechanism is described for facilitating dynamic and remote memory collaboration at computing devices according to one embodiment of the invention. A method of embodiments of the invention includes dynamically classifying a computing device of a plurality of computing devices as a memory server, where the plurality of computing devices are coupled to each other over a network. The method may further include offering, by the memory server, of memory to be used by one or more of the plurality of computing devices classified as one or more memory clients, and remotely granting, by the memory server, of the memory to the one or more memory clients.

Abstract: A system and method for maintaining connectivity between a host system running an Always-On-Always-Connected (AOAC) application and an associated remote application server includes determining a timing interval Ti for sending keep-alive messages. The timing interval Ti may be determined by selecting a value for a timeout (Ti) to a value between a maximum timeout (Tmax) and a minimum timeout (Tmin), transmitting a keep-alive message, at an interval based on Ti, across a network connection between a client platform running an Always-On-Always-Connected (AOAC) application and a remote application server associated with the AOAC application, checking a status of the network connection, increasing the value for Tmin if the network connection is still active and decreasing the value for Tmax if the network connection has been dropped.

Abstract: An embodiment may include circuitry to determine whether to issue at least one credit to at least one sender of at least one packet. The credit(s) may be to grant permission to the at least one sender to issue the at least one packet to at least one receiver of the at least one packet. The determination of whether to issue the credit(s) may be based, at least in part, upon whether a time in which the at least one receiver is in a relatively lower power state prior to issuance of the credit(s) is at least sufficient to provide at least a predetermined amount of reduction in power consumption. The relatively lower power state may be relative to a relatively higher power state of the at least one receiver that prevails at the issuance of the credit(s). Additionally or alternatively, the circuitry may be to receive such credit(s).

Abstract: An embodiment may include network controller to be comprised in a first node. The node may be communicatively coupled to a network and may include a host processor to execute an operating system environment. The operating system environment may include, at least in part, a communication protocol stack and an application. The circuitry may receive, at least in part, a packet from the network. The packet may include, at least in part, a header and payload. At least one portion of the payload may be associated with the application. The circuitry may issue at least one portion of the header to the stack. The circuitry may issue the at least one portion of the payload to a destination device in a manner that by-passes involvement of the stack. The destination device may be specified, at least in part, by the application. Many alternatives, variations, and modifications are possible.

Abstract: A system and method for maintaining connectivity between a host system running an Always-On-Always-Connected (AOAC) application and an associated remote application server. The system further includes circuitry configured to establish a communication link between the host system and the remote application server. The circuitry is configured periodically transmit keep-alive messages to the remote application server after the host system transitions to and remains in a low-power state. The keep-alive messages are configured to maintain connectivity and presence of the AOAC application with the remote application server while the host system is in the low-power state.

Abstract: The present invention relates to platform power management. In some implementations, platform tasks, that require servicing by a host processor, may be serviced in groups to create longer or more idle periods to enable the host processor to be in lower power consuming states more often.

Abstract: Systems and methods may provide for determining an absolute energy break-even time for a first low power state with respect to a current state of a system. A relative energy break-even time may also be determined for the first low power state with respect to a second low power state based on at least in part the absolute energy break-even time. In addition, an operating state may be selected for the system based on at least in part the relative energy break-even time.