Technologies for an accelerator interface over Ethernet are disclosed. In the illustrative embodiment, a network interface controller of a compute device may receive a data packet. If the network interface controller determines that the data packet should be pre-processed (e.g., decrypted) with a remote accelerator device, the network interface controller may encapsulate the data packet in an encapsulating network packet and send the encapsulating network packet to a remote accelerator device on a remote compute device. The remote accelerator device may pre-process the data packet (e.g., decrypt the data packet) and send it back to the network interface controller. The network interface controller may then send the pre-processed packet to a processor of the compute device.

Technologies for managing network statistic counters include a network interface controller (NIC) of a computing device configured to identify a statistic counter of and a software consumer associated with a received network packet and identify an active counter page as a function of the identified software consumer. The NIC is further configured to read a value of the statistic counter stored at a counter memory address of a corresponding counter identifier entry of the identified active counter page, increment a read value of the statistic counter, and write the incremented value of the statistic counter back to the counter memory address. Additionally, in response to detecting a notification triggering event, generating a notification message that includes a present value of the statistic counter and a present value of each of the other statistic counters of the active counter page, and transmit the generated notification message to the software consumer. Other embodiments are described herein.

Technologies for processing network packets in an agent-mesh architecture include a network interface controller (NIC) of a computing device configured to write, by a network fabric interface of a memory fabric of the NIC, a received network packet to the memory fabric in a distributed fashion. The network fabric interface is configured to send an event message indicating the received network packet to a packet processor communicatively coupled to the memory fabric. The packet processor is configured to read, in response to having received the generated event message, at least a portion of the received network packet from the memory fabric, identify an agent of the NIC for additional processing of the received network packet, generate a network packet received event message indicating the received network packet is available for processing, and transmit the network packet received event message to the identified agent. Other embodiments are described herein.

Technologies for processing network packets by a network interface controller (NIC) of a computing device include a network interface, a packet processor, and a controller device of the NIC, each communicatively coupled to a memory fabric of the NIC. The packet processor is configured to receive an event message from the memory fabric and transmit a message to the controller device, wherein the message indicates the network packet has been received and includes the memory fabric location pointer. The controller device is configured to fetch at least a portion of the received network packet from the memory fabric, write an inbound descriptor usable by one or more on-die cores of the NIC to perform an operation on the fetched portion, and restructure the network packet as a function of an outbound descriptor written by the on-die cores subsequent to performing the operation. Other embodiments are described herein.

In the conventional eFPGAs, there have been two challenges: the first one being size reduction of the logic cells to improve the implemented logic density, and the second one being minimization of the speed difference with the ASIC. According to the present embodiment, there is provided a method for configuring a programmable logic circuit represented by a gate-level netlist, wherein this is done by assigning the gate-level netlist to a 4-input, 3-output combinational logic cell, which is composed of a combination of 3 of 2-input (m-input) basic logic cells, wherein the combinational logic cell covers 3(n) nodes constituting a graph of the netlist.

Embodiments of the invention relate to a metal configurable hybrid memory for use in integrated circuit designs for implementation in structured ASIC or similar platforms utilizing a base cell or standard cell. In accordance with certain aspects, a hybrid memory according to embodiments of the invention utilizes a fixed custom memory core and a customizable peripheral set of base cells. In accordance with these and further aspects, the hybrid memory can be specified using a macro, in which certain memory features (e.g. ECC, etc.) are implemented using the customizable peripheral set of base cells, and which may be selected or omitted from the design by the user. This enables the overall logic use for the memory to be optimized for a user's particular design. Unused logic in the customizable peripheral set of base cells can thus be freed for top-level logic use, thereby optimizing the design according to a user's functional and dimensional requirements and minimizing unnecessary waste of silicon area and power.

Technologies for offloading acceleration task scheduling operations to accelerator sleds include a compute device to receive a request from a compute sled to accelerate the execution of a job, which includes a set of tasks. The compute device is also to analyze the request to generate metadata indicative of the tasks within the job, a type of acceleration associated with each task, and a data dependency between the tasks.

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.

Technologies for offloading acceleration task scheduling operations to accelerator sleds include a compute device to receive a request from a compute sled to accelerate the execution of a job, which includes a set of tasks. The compute device is also to analyze the request to generate metadata indicative of the tasks within the job, a type of acceleration associated with each task, and a data dependency between the tasks. Additionally the compute device is to send an availability request, including the metadata, to one or more micro-orchestrators of one or more accelerator sleds communicatively coupled to the compute device. The compute device is further to receive availability data from the one or more micro-orchestrators, indicative of which of the tasks the micro-orchestrator has accepted for acceleration on the associated accelerator sled. Additionally, the compute device is to assign the tasks to the one or more micro-orchestrators as a function of the availability data.