Abstract: Methods and apparatus for mapping memory allocation to DRAM dies of a stacked memory modules are described herein. Memory address ranges in a module employing 3DS (three dimensional stacked) DRAMs (Dynamic Random Access Memories) comprising stacked DRAM dies are mapped to DRAM dies in the module based on a layer of the DRAM dies, where dies in different layers have different thermal dissipation characteristic. Chunks of the memory address range are allocated to software entities such as virtual machines (VMs) and/or applications based on a memory access rate of the VMs/applications and the thermal dissipation characteristics of the DRAM die layers, wherein VMs/applications with higher memory access rate are allocated memory on DRAM dies with higher thermal dissipation. In one aspect, memory ranks are associated with respective die layers. In response to detection of change in access rates, memory may be migrated between ranks. Interleaving at multiple levels is also supported.

Abstract: A memory system includes a first set of memory devices, a second set of memory devices, and a memory controller circuit system. The memory controller circuit system groups a first one of the memory devices in each of the first and the second sets into a first virtual memory rank based on eye margins of first data signals sampled by the first virtual memory rank. The memory controller circuit system groups a second one of the memory devices in each of the first and the second sets into a second virtual memory rank based on eye margins of second data signals sampled by the second virtual memory rank. The memory controller circuit system accesses the memory devices in the first virtual memory rank separately from the memory devices in the second virtual memory rank during data access operations.

Abstract: An apparatus comprising a first processor core to execute a first instance of an application; a second processor core to execute a second instance of the application concurrent with the execution of the first instance of the application; and processing circuitry to direct an interrupt to the first processor core based on an indication that an execution state of the first processor core is ahead of an execution state of the second processor core.

Abstract: Techniques to shared system memory across nodes in a system. Circuitry is arranged to provide a mechanism to share a memory region of a memory maintained at a first host CPU at a first node across multiple other host CPUs at multiple other nodes using various links and protocols described in one or more revisions of the Compute Express Link (CXL) specification.

Abstract: Methods and apparatus relating to enhancing diagnostic capabilities of computing systems by combining variable patrolling Application Program Interface (API) and comparison mechanism of variables are described. In one embodiment, a first processor core executes a first instance of a workload to generate a first set of safety variables. A second processor core executes a second instance of the workload to generate a second set of safety variables. A third processor core generates a signal in response to comparison of the first set of safety variables and the second set of safety variables. Other embodiments are also disclosed and claimed.

Abstract: Disclosed herein are mechanisms and methods for reducing power consumed by various DRAM technologies (e.g., high-capacity DRAM and/or 3D DRAM) which may impact battery life of the platform. These mechanisms and methods may opportunistically reduce the power consumed by DRAM by inhibiting periodic refresh commands to memory ranks that are not in-use. Since these mechanisms and methods may be based on enhancements to memory controllers, they may accordingly be operating system (OS) agnostic.

Abstract: Methods and apparatus for mapping memory allocation to DRAM dies of a stacked memory modules are described herein. Memory address ranges in a module employing 3DS (three dimensional stacked) DRAMs (Dynamic Random Access Memories) comprising stacked DRAM dies are mapped to DRAM dies in the module based on a layer of the DRAM dies, where dies in different layers have different thermal dissipation characteristic. Chunks of the memory address range are allocated to software entities such as virtual machines (VMs) and/or applications based on a memory access rate of the VMs/applications and the thermal dissipation characteristics of the DRAM die layers, wherein VMs/applications with higher memory access rate are allocated memory on DRAM dies with higher thermal dissipation. In one aspect, memory ranks are associated with respective die layers. In response to detection of change in access rates, memory may be migrated between ranks. Interleaving at multiple levels is also supported.

Abstract: A memory system includes a first set of memory devices, a second set of memory devices, and a memory controller circuit system. The memory controller circuit system groups a first one of the memory devices in each of the first and the second sets into a first virtual memory rank based on eye margins of first data signals sampled by the first virtual memory rank. The memory controller circuit system groups a second one of the memory devices in each of the first and the second sets into a second virtual memory rank based on eye margins of second data signals sampled by the second virtual memory rank. The memory controller circuit system accesses the memory devices in the first virtual memory rank separately from the memory devices in the second virtual memory rank during data access operations.

Abstract: An apparatus comprising a first processor core to execute a first instance of an application; a second processor core to execute a second instance of the application concurrent with the execution of the first instance of the application; and processing circuitry to direct an interrupt to the first processor core based on an indication that an execution state of the first processor core is ahead of an execution state of the second processor core.

Abstract: Disclosed herein are mechanisms and methods for reducing power consumed by various DRAM technologies (e.g., high-capacity DRAM and/or 3D DRAM) which may impact battery life of the platform. These mechanisms and methods may opportunistically reduce the power consumed by DRAM by inhibiting periodic refresh commands to memory ranks that are not in-use. Since these mechanisms and methods may be based on enhancements to memory controllers, they may accordingly be operating system (OS) agnostic.

Abstract: Methods and apparatus for initializing memory using a hardware engine for minimizing boot time. Booting of firmware in a computer system including one or more memory devices and a processor System on a Chip (SoC) including a central processing unit (CPU) having a plurality of cores and a memory controller coupled to the one or more memory devices is initialized. In parallel with at least a portion of booting the firmware, a plurality of memory ranges in at least one memory devices is initialized using at least one scrub engine. The at least one scrub engine may be embedded in the memory controller or external to the memory controller. Memory devices and ranges within those devices may continue to be initialized in parallel with booting an operating system following firmware booting. The scrub engine includes one or more registers or embedded memory used to program/store descriptor chains describing memory ranges to be initialized.

Abstract: Disclosed herein are mechanisms and methods for reducing power consumed by various DRAM technologies (e.g., high-capacity DRAM and/or 3D DRAM) which may impact battery life of the platform. These mechanisms and methods may opportunistically reduce the power consumed by DRAM by inhibiting periodic refresh commands to memory ranks that are not in-use. Since these mechanisms and methods may be based on enhancements to memory controllers, they may accordingly be operating system (OS) agnostic.

Abstract: Technologies for reliable software execution include a computing device having a memory that includes multiple ranks. The computing device trains the ranks of the memory and determines a consolidated memory score for each rank. Each consolidated memory score is indicative of a margin of the corresponding rank. The computing device identifies a higher-margin address range using the consolidated memory scores. The higher-margin memory address range is mapped to a higher-margin memory rank. The computing device loads high-priority software into the higher-margin memory address range. The high-priority software may include an operating system or a critical application. A pre-boot firmware environment may publish the consolidated memory scores to a higher-level software component, such as the operating system. The pre-boot firmware environment may map a predetermined address range to the higher-margin memory rank. A critical application may request to be loaded into a higher-margin address range.

Abstract: Methods and apparatus relating to enhancing diagnostic capabilities of computing systems by combining variable patrolling Application Program Interface (API) and comparison mechanism of variables are described. In one embodiment, a first processor core executes a first instance of a workload to generate a first set of safety variables. A second processor core executes a second instance of the workload to generate a second set of safety variables. A third processor core generates a signal in response to comparison of the first set of safety variables and the second set of safety variables. Other embodiments are also disclosed and claimed.

Abstract: Technologies for reliable software execution include a computing device having a memory that includes multiple ranks. The computing device trains the ranks of the memory and determines a consolidated memory score for each rank. Each consolidated memory score is indicative of a margin of the corresponding rank. The computing device identifies a higher-margin address range using the consolidated memory scores. The higher-margin memory address range is mapped to a higher-margin memory rank. The computing device loads high-priority software into the higher-margin memory address range. The high-priority software may include an operating system or a critical application. A pre-boot firmware environment may publish the consolidated memory scores to a higher-level software component, such as the operating system. The pre-boot firmware environment may map a predetermined address range to the higher-margin memory rank. A critical application may request to be loaded into a higher-margin address range.