一种堆叠芯片
    1.
    发明申请
    一种堆叠芯片 审中-公开

    公开(公告)号:WO2023030051A1

    公开(公告)日:2023-03-09

    申请号:PCT/CN2022/113699

    申请日:2022-08-19

    IPC分类号: G06F15/76

    摘要: 一种堆叠芯片,其中堆叠芯片包括:第一可编程门阵列组件,第一可编程门阵列组件包括第一接口模块,第一接口模块嵌入于第一可编程门阵列组件内,第一接口模块包括第一键合引出区域;第一存储阵列组件,设置有第二键合引出区域;第一键合引出区域、第二键合引出区域键合连接,以将第一可编程门阵列组件以及第一存储阵列组件上的互连信号连接在一起。实现存储访问的高带宽、低功耗的目的。

    向量指令的识别方法、装置、电子设备及计算机可读存储介质

    公开(公告)号:WO2022206969A1

    公开(公告)日:2022-10-06

    申请号:PCT/CN2022/084890

    申请日:2022-04-01

    IPC分类号: G06F9/30 G06F15/76

    摘要: 本公开涉及一种向量指令的识别方法、装置、电子设备和计算机可读存储介质。其中该方法包括:获取标量指令;根据向量指令维度映射规则以及指令配置表将所述标量指令映射为向量指令;其中,所述向量指令维度映射规则包括将标量指令映射为向量指令的规则;所述指令配置表包括向量指令集合;根据向量指令维度有效性规则判断所述向量指令的有效性;其中,所述向量指令维度有效性规则包括验证指令映射结果的有效性的规则;如果所述向量指令有效,则所述向量指令识别成功。上述方法通过向量指令维度映射规则、指令配置表以及向量指令维度有效性规则识别向量指令,解决了将标量指令转换为向量指令时复杂、通用性不高且无法区分向量指令的技术问题。

    ADVANCED QUANTUM PROCESSING SYSTEMS AND METHODS

    公开(公告)号:WO2022032331A1

    公开(公告)日:2022-02-17

    申请号:PCT/AU2021/050869

    申请日:2021-08-09

    IPC分类号: G06N10/00 G06F15/76

    摘要: Quantum processing devices and methods for shuttling qubits between a pair of processing elements are disclosed. In particular, a disclosed method for shuttling a qubit from a first processing element to a second processing element in a quantum processing device comprising a plurality of processing elements, includes: applying an optimal bias configuration between the first processing element and the second processing element to shuttle the qubit from the first processing element to the second processing element in a manner that minimizes the time spent by the qubit in one or more state transition points between the first processing element and the second processing element.

    一种计算装置、集成电路芯片、板卡、设备和计算方法

    公开(公告)号:WO2022001438A1

    公开(公告)日:2022-01-06

    申请号:PCT/CN2021/094467

    申请日:2021-05-18

    IPC分类号: G06F15/76

    摘要: 本披露公开了一种计算装置、集成电路芯片、板卡、设备和方法。该计算装置可以包括在组合处理装置中,该组合处理装置还可以包括接口装置和其他处理装置。该计算装置与其他处理装置进行交互,共同完成用户指定的计算操作。组合处理装置还可以包括存储装置,该存储装置分别与计算装置和其他处理装置连接,用于存储该计算装置和其他处理装置的数据。本披露的方案可以利用表征大位宽数据的至少两个小位宽数据来执行运算处理,以使得处理器的处理能力不受位宽的影响。摘要附图:图7

    METHOD AND APPARATUS FOR COMPILING COMPUTATION GRAPHS INTO AN INTEGRATED CIRCUIT

    公开(公告)号:WO2020087072A1

    公开(公告)日:2020-04-30

    申请号:PCT/US2019/058363

    申请日:2019-10-28

    申请人: TENSIL AI COMPANY

    IPC分类号: G06F17/50 G06F15/76 G06F15/82

    摘要: Disclosed are systems and methods for a compiler, which can receive a computation workload, and a description of the computation graph of the workload and compile a circuit layout of the workload. In one embodiment, an RTL generator assigns the node operations of the computation graph to a first or second type. In the first type, the workload is loaded and processed in tiles equal to a compute filter width. In the second type, the workload is loaded in tiles larger in size than the width of the compute filter, allowing the compute filter to process more operations in parallel and reach the data needed for the underlying operations more efficiently.

    Processor Instructions to Accelerate FEC Encoding and Decoding
    7.
    发明申请
    Processor Instructions to Accelerate FEC Encoding and Decoding 审中-公开
    处理器指令加速FEC编码和解码

    公开(公告)号:WO2017117116A1

    公开(公告)日:2017-07-06

    申请号:PCT/US2016/068692

    申请日:2016-12-27

    IPC分类号: G06F9/30 G06F9/38 G06F15/76

    摘要: Various embodiments are described of a system for improved processor instructions for a software-configurable processing element. In particular, various embodiments are described which accelerate functions useful for FEC encoding and decoding. In particular, the processing element may be configured to implement one or more instances of the relevant functions in response to receiving one of the processor instructions. The processing element may later be reconfigured to implement a different function in response to receiving a different one of the processor instructions. Each of the disclosed processor instructions may be implemented repeatedly by the processing element to repeatedly perform one or more instances of the relevant functions with a throughput approaching one or more solutions per clock cycle.

    摘要翻译: 描述用于软件可配置处理元件的改进处理器指令的系统的各种实施例。 具体而言,描述了加速对FEC编码和解码有用的功能的各种实施例。 特别地,处理元件可以被配置为响应于接收处理器指令之一来实现相关功能的一个或多个实例。 稍后可以将处理元件重新配置为响应于接收不同的一个处理器指令而实现不同的功能。 所公开的处理器指令中的每一个可以由处理元件重复地实现,以每个时钟周期接近一个或多个解决方案的吞吐量来重复执行相关功能的一个或多个实例。

    SCALABLE POLYLITHIC ON-PACKAGE INTEGRATABLE APPARATUS AND METHOD
    8.
    发明申请
    SCALABLE POLYLITHIC ON-PACKAGE INTEGRATABLE APPARATUS AND METHOD 审中-公开
    可伸缩多晶石墨包装可集成装置和方法

    公开(公告)号:WO2017099908A1

    公开(公告)日:2017-06-15

    申请号:PCT/US2016/059962

    申请日:2016-11-01

    申请人: INTEL CORPORATION

    IPC分类号: G06F15/76

    摘要: Described is an apparatus which comprises: a first die including: a processing core; a crossbar switch coupled to the processing core; and a first edge interface coupled to the crossbar switch; and a second die including: a first edge interface positioned at a periphery of the second die and coupled to the first edge interface of the first die, wherein the first edge interface of the first die and the first edge interface of the second die are positioned across each other; a clock synchronization circuit coupled to the second edge interface; and a memory interface coupled to the clock synchronization circuit.

    摘要翻译: 描述了一种装置,其包括:第一模具,其包括:处理核心; 耦合到处理核心的交叉开关; 以及耦合到所述纵横开关的第一边缘接口; 以及第二模具,所述第二模具包括:第一边缘界面,所述第一边缘界面定位在所述第二模具的外围并且耦合到所述第一模具的所述第一边缘界面,其中所述第一模具的所述第一边缘界面和所述第二模具的所述第一边缘界面被定位 相互之间; 时钟同步电路,其耦合到所述第二边缘接口; 以及耦合到时钟同步电路的存储器接口。

    AUTOMATED COMPUTER SYSTEM AND METHOD
    9.
    发明申请
    AUTOMATED COMPUTER SYSTEM AND METHOD 审中-公开
    自动计算机系统及方法

    公开(公告)号:WO2017010939A1

    公开(公告)日:2017-01-19

    申请号:PCT/SG2016/050322

    申请日:2016-07-11

    IPC分类号: G06F17/00 G06F15/76

    摘要: Embodiments disclosed include computer automated systems and methods for aggregating data from a plurality of data sources, such as proxy devices, legacy protocols, devices, applications, machines, sensors, things across locations and user types, or device clouds among devices and applications. The aggregated data is then normalized and the normalized data is analyzed. The analyzing is based on a correlated event or events, a correlated condition or conditions, and a correlated trend or trends across the plurality of data sources. And based on the analyzed data, relevant aggregated and normalized data is combined and displayed in a display compatible format. Additionally, user needs are determined based on the analyzed aggregated, normalized data. The user need comprises a need for an item or items comprising at least one of a service, a product, and an upgrade of hardware or software components. Further a provider from a plurality of providers is determined based on the determined user need, and finally a need fulfillment transaction between the user and the provider is initiated.

    摘要翻译: 所公开的实施例包括计算机自动化系统和方法,用于从多个数据源(例如代理设备,传统协议,设备,应用,机器,传感器,跨位置的事物和用户类型或设备和应用中的设备云)聚合数据。 然后对聚合数据进行归一化,并分析归一化数据。 该分析基于相关事件或事件,相关条件或多个数据源中的相关趋势或趋势。 并根据分析数据,将相关的聚合和归一化数据组合并以显示兼容格式显示。 另外,基于所分析的聚合的标准化数据来确定用户需求。 用户需要包括对包括服务,产品和硬件或软件组件的升级中的至少一个的项目或项目的需求。 此外,来自多个提供者的提供者基于所确定的用户需求来确定,并且最终启动用户和提供者之间的需求履行交易。