Abstract: Deployment of arrangements of computing components coupled over a communication fabric are presented herein. In one example, a method includes detecting first computing components communicatively coupled to a first communication fabric having a first communication fabric type, and detecting second computing components communicatively coupled to a second communication fabric having a second communication fabric type. The method also includes receiving user commands to form compute units among a pool of computing components comprising the first computing components and the second computing components. Based at least on the user commands, the method includes forming the compute units for use by one or more users.

Abstract: Disaggregated computing architectures, platforms, and systems are provided herein. In one example, a method includes instructing a communication fabric to establish a first logical partition in the communication fabric that includes a first processing device and a memory device, and directing transfer of configuration data for storage by the memory device over the first logical partition. After transfer of the configuration data, the method includes instructing the communication fabric to remove the first logical partition in the communication fabric, where the configuration data remains stored by the memory device after removal of the first logical partition. The method also includes instructing the communication fabric to establish a second logical partition in the communication fabric that includes at least a second processing device and the memory device comprising the configuration data, where the second processing device is operated using the configuration data.

Abstract: Deployment of arrangements of computing components coupled over communication fabrics are presented herein. In one example, a method includes forming a compute unit to include at least a selected disaggregated computing component and at least a selected converged computing element. The selected disaggregated computing component is added to the compute unit from a pool of disaggregated computing components having a granularity comprising the disaggregated computing components arbitrarily and separately coupled to a communication fabric. The selected converged computing element is added to the compute unit from a pool of converged computing elements having a granularity comprising sets of computing components with fixed relationships to associated processors.

Abstract: Enhanced apparatuses, systems, and techniques for coupling network-linked peripheral devices into host computing devices is presented. A method includes, over a network interface of a host device, obtaining an indication of a peripheral device available for associating with the host device. Based on the indication, the method includes initiating instantiation of the peripheral device into a Peripheral Component Interconnect Express (PCIe) subsystem of the host device by at least emulating behavior of the peripheral device over the network interface as a PCIe peripheral device coupled locally to the host system.

Abstract: Communication fabric-coupled computing architectures, platforms, and systems are provided herein. In one example, an apparatus includes a management entity configured to establish a compute unit comprising components from among a plurality of physical computing components by at least instructing a communication fabric communicatively coupling the plurality of physical computing components to establish logical isolation within the communication fabric to form the compute unit. Responsive to an indication of a change in workload associated with at least a software component deployed to a processing element of the compute unit, the management entity is configured to adjust the logical isolation to alter a quantity of the plurality of physical computing components in the compute unit in accordance with the change in the workload.

Abstract: Computing architectures, platforms, and systems are provided herein. In one example, system is provided. The system includes a management processor configured to initiate a communication arrangement between a first endpoint device coupled to a communication fabric and a second endpoint device coupled to the communication fabric. The communication arrangement is configured to redirect a transfer from the first endpoint device based on an address corresponding to an address range of the second endpoint device without passing the transfer through a host processor coupled to the communication fabric that executes an application initiating the transfer.

Abstract: Enhanced apparatuses, systems, and techniques for coupling network-linked peripheral devices into host computing devices is presented. A method includes, over a network interface of a host device, obtaining an indication of a peripheral device available for associating with the host device. Based on the indication, the method includes initiating instantiation of the peripheral device into a Peripheral Component Interconnect Express (PCIe) subsystem of the host device by at least emulating behavior of the peripheral device over the network interface as a PCIe peripheral device coupled locally to the host system.

Abstract: Emulated telemetry interfaces for host processors and management processors coupled over communication fabrics are presented herein. In one example, an apparatus includes a monitoring function executed by a host processor configured to determine telemetry related to operation of at least the host processor. A driver function executed by the host processor is configured to emulate operation of a network interface to an operating system of the host processor for transfer of communications comprising at least the telemetry to a management processor over a communication fabric different than the network interface, where the host processor and the management processor are communicatively coupled to the communication fabric.

Abstract: Peer-to-peer arrangements between endpoint devices are provided herein. A method includes establishing synthetic devices representing endpoint devices in an address domain associated with a host processor, where the endpoint devices have a different address domain than the host processor. The method also includes forming a peer arrangement between the endpoint devices such that data transfers between the endpoint devices in the different address domain can be initiated by the host processor interfacing with the synthetic devices.

Abstract: Enhanced apparatuses, systems, and techniques for coupling network-linked peripheral devices into host computing devices is presented. A method includes, over a network interface of a host device, obtaining an indication of a peripheral device available for associating with the host device. Based on the indication, the method includes initiating instantiation of the peripheral device into a Peripheral Component Interconnect Express (PCIe) subsystem of the host device by at least emulating behavior of the peripheral device over the network interface as a PCIe peripheral device coupled locally to the host system.

Abstract: Enhanced apparatuses, systems, and techniques are presented for coupling remote peripheral devices located at a host device, as local peripheral devices of a client device. A method includes receiving an indication of a peripheral device, located at a first computing device, that is available for data processing, and receiving a request from a second computing device to access the peripheral device. Based on the request, the method may include directing the second computing device to emulate the peripheral device as a local device installed into the second computing device, and routing data traffic from the second computing device to the first computing device for processing by the peripheral device.

Abstract: Peer-to-peer arrangements between graphics processing units (GPUs) are provided herein. A method includes establishing synthetic devices representing GPUs in an address domain associated with a host processor, where the GPUs have a different address domain than the host processor. The method also includes forming a peer arrangement between the GPUs such that data transfers between the GPUs in the different address domain can be initiated by the host processor interfacing with the synthetic devices.

Abstract: Computing architectures, platforms, and systems are provided herein. In one example, system is provided. The system includes a first processor configured to initiate a communication arrangement between a first peripheral component interconnect express (PCIe) device and a second PCIe device. The communication arrangement is configured to detect transfers from the first PCIe device to one or more addresses corresponding to an address range of the second PCIe device, and redirect the transfers to the second PCIe device without passing the transfers through a second processor that initiates the transfers.

Abstract: Software configuration deployment techniques for disaggregated computing architectures, platforms, and systems are provided herein. In one example, a method includes presenting a user interface configured to receive instructions related to deployment of software to compute units, and receiving user selections of a software element for deployment to a compute unit comprising a processing element and a storage element.

Abstract: A method of programming a fabric-coupled FPGA device includes instructing a communication fabric to establish a first logical partition segregating a first set of communication switch ports within the communication fabric that includes the management processor and a target device comprising a memory device configured to program a field-programmable gate array (FPGA). The method includes directing configuration data configured to program the FPGA to the target device over the first logical partition. The method includes instructing the communication fabric to remove the first logical partition in the communication fabric, wherein the configuration data remains stored at the target device after removal of the first logical partition. The method includes instructing the communication fabric to establish a second logical partition segregating a second set of communication switch ports within the communication fabric that includes a selected device and the target device that stores the configuration data.

Abstract: Deployment of arrangements of physical computing components coupled over a communication fabric are presented herein. In one example, a method includes presenting, to a workload manager, a target machine capable of receiving execution jobs from the workload manager. The target machine has a network state and comprises a selection of computing components. The method also includes receiving a job issued by the workload manager that is directed to the target machine. Based on properties of the job, the method includes determining resource requirements for handling the job, forming a composed machine comprising physical computing components that support the resource requirements of the job, transferring the network state of the target machine to the composed machine and indicating the network state of the composed machine to the workload manager, and initiating execution of the job on the composed machine.

Abstract: Disaggregated computing architectures, platforms, and systems are provided herein. In one example, a method of operating a data processing system is provided. The method includes instructing a PCIe fabric communicatively coupling a plurality of physical computing components including PCIe devices and one or more PCIe switches to establish a first PCIe communication path between the management processor and a target PCIe device. The method also includes directing at least configuration data to the target PCIe device using the first PCIe communication path and instructing the PCIe fabric to remove the first PCIe communication path between the management processor and the target PCIe device. The method also includes instructing the PCIe fabric to establish a second PCIe communication path between a selected PCIe device and the target PCIe device configured according to the configuration data.

Abstract: Peer-to-peer arrangements between graphics processing units (GPUs) are provided herein. A method includes establishing synthetic devices representing GPUs in an address domain associated with a host processor, where the GPUs have a different address domain than the host processor. The method also includes forming a peer arrangement between the GPUs such that data transfers between the GPUs in the different address domain can be initiated by the host processor interfacing with the synthetic devices.

Abstract: Emulated telemetry interfaces for host processors and management processors coupled over communication fabrics are presented herein. In one example, an apparatus includes a monitoring function executed by a host processor configured to determine telemetry related to operation of at least the host processor. A driver function executed by the host processor is configured to emulate operation of a network interface to an operating system of the host processor for transfer of communications comprising at least the telemetry to a management processor over a communication fabric different than the network interface, where the host processor and the management processor are communicatively coupled to the communication fabric.

Abstract: Disaggregated computing architectures, platforms, and systems are provided herein. In one example, a method includes instructing a communication fabric to establish a first logical partition in the communication fabric that includes a first processing device and a memory device, and directing transfer of configuration data for storage by the memory device over the first logical partition. After transfer of the configuration data, the method includes instructing the communication fabric to remove the first logical partition in the communication fabric, where the configuration data remains stored by the memory device after removal of the first logical partition. The method also includes instructing the communication fabric to establish a second logical partition in the communication fabric that includes at least a second processing device and the memory device comprising the configuration data, where the second processing device is operated using the configuration data.