A recording medium stores a program for a process including: receiving operation information for determining whether to perform offload from a first integrated circuit to a second integrated circuit; performing first judgment of determining that the offload is possible when resolution is equal to or higher than a threshold; performing second judgment of determining that the offload is possible when it is determined in accordance with the operation information that access to a memory included in the second integrated circuit is continuous in the operation; performing third judgment of determining that the offload is possible when it is determined in accordance with the operation information that the operation is able to be performed in a parallelized manner; and performing determination processing of determining whether the offload is possible in accordance with a judgment by the first judgment, a judgment by the second judgment, and a judgment by the third judgment.

Systems and methods for utilizing a defect map to configure an automata processor in order to avoid defects when configuring the automata processor. A system includes automata processor having a state machine lattice. The system also includes a non-volatile memory having a defect map stored thereon and indicating logical defects found on the automata processor. By including the defect map, a compiler may access the defect map to map out defects in the automata processor during configuring to avoid such defects.

A computer system includes a processor and a memory. The processor is located on a first circuit board having a first connector. The memory is located on a second circuit board having a second connector. The first circuit board and the second board are physically separated from each other but connect to each other through the connector. The processor and the memory are communicated to each other based on a differential signaling scheme.

The embodiments described herein describe technologies for memory systems. One implementation of a memory system includes a motherboard substrate with multiple module sockets, at least one of which is populated with a memory module. A first set of data lines is disposed on the motherboard substrate and coupled to the module sockets. The first set of data lines includes a first subset of point-to-point data lines coupled between a memory controller and a first socket and a second subset of point-to-point data lines coupled between the memory controller and a second socket. A second set of data lines is disposed on the motherboard substrate and coupled between the first socket and the second socket. The first and second sets of data lines can make up a memory channel.

The present invention relates to cache coherent node controllers for scale-up shared memory systems. In particular it is disclosed a computer system at least comprising a first group of CPU modules connected to at least one first FPGA Node Controller configured to execute transactions directly or through a first interconnect switch to at least one second FPGA Node Controller connected to a second group of CPU modules running a single instance of an operating system.

A multi-direction connectable electronic module includes a circuit board, including a top surface, a bottom surface, and at least one side; and a plurality of connectors connected to the circuit board, each including a lateral magnetic connector, a shell, a longitudinal inter-locking part, and a lateral inter-locking part. The lateral inter-locking part is configured to connect with a first electronic building block along the lateral direction. The longitudinal inter-locking part is configured to stack with a second electronic building block along the longitudinal direction. The lateral magnetic connector is configured to magnetically connect with the first electronic building block. A plurality of through holes are formed on the shell. A lateral pin connector disposed on the at least one side of the circuit board includes a plurality of pins located at positions corresponding to the plurality of through holes, and is configured to electrically connect the first electronic building block.

A multichip package having a main die coupled to one or more daughter dies is provided. The main die may include embedded universal interface blocks (UIB) each of which can be used to interface with a corresponding daughter die to support high bandwidth parallel or serial communications. Each UIB may include an integrated processor subsystem and associated pattern sequencing logic to perform interface initialization and margining operations. Each UIB may perform simultaneous accesses to a daughter die across one or more channels. Each UIB may also include multiple phase-locked loop circuits for providing different clock signals to different portions of the UIB and a 2 clock phase generation logic. Each UIB may include multiple IO modules, each of which may optionally include its own duty cycle correction circuit. Each IO module may include buffer circuits, each of which may have a de-emphasis control logic for adjusting buffer drive strength.

A method is provided for processing on an acceleration component a deep neural network. The method includes configuring the acceleration component to perform forward propagation and backpropagation stages of the deep neural network. The acceleration component includes an acceleration component die and a memory stack disposed in an integrated circuit package. The memory stack has a memory bandwidth greater than about 50 GB/sec and a power efficiency of greater than about 20 MB/sec/mW.

Random access memory (RAM) is attached to an input/output (I/O) controller of a chipset (e.g., on a motherboard). This chipset attached RAM is optionally used as part of a tiered storage solution with other tiers including, for example, nonvolatile memory (e.g., a solid state drive (SSD)) or a hard disk drive. The chipset attached RAM is separate from the system memory, allowing the chipset attached RAM to be used to speed up access to frequently used data stored in the tiered storage solution without reducing the amount of system memory available to an operating system running on the one or more processing units.

A computing device can flexibly connect bidirectional processor interconnect ports (BPIPs). An exemplary computing device includes a motherboard structure, a first processor, a second processor, and a plurality of connectors disposed on the motherboard structure. The first processor and the second processor can each have at least three BPIPs. A first and second of BPIPs of the first processor can be connected to a first and second BPIPs of the second processor. A third BPIP of the first processor and a third BPIP of the second processor can be connected to a first one and a second one of the plurality of connectors. The plurality of connectors can be connected to a computing card. In some examples, the computing device includes a switching element to selectively couple the connectors to any other element in the computing device.