ADAPTIVE MATRIX MULTIPLICATION ACCELERATOR FOR MACHINE LEARNING AND DEEP LEARNING APPLICATIONS

    公开(公告)号:US20230041850A1

    公开(公告)日:2023-02-09

    申请号:US17967733

    申请日:2022-10-17

    Abstract: An adaptive matrix multiplier. In some embodiments, the matrix multiplier includes a first multiplying unit a second multiplying unit,a memory load circuit, and an outer buffer circuit. The first multiplying unit includes a first inner buffer circuit and a second inner buffer circuit, and the second multiplying unit includes a first inner buffer circuit and a second inner buffer circuit. The memory load circuit is configured to load data from memory, in a single burst of a burst memory access mode, into the first inner buffer circuit of the first multiplying unit; and into the first inner buffer circuit of the second multiplying unit.

    Embedded reference counter and special data pattern auto-detect

    公开(公告)号:US11079954B2

    公开(公告)日:2021-08-03

    申请号:US16180002

    申请日:2018-11-04

    Abstract: A deduplication memory system includes a virtual memory space, a physical memory space and a memory manager. The memory manager generates a user data entry that is stored in the physical memory space. The user data entry represents a unique user data of a predetermined granularity appearing in the virtual memory space, and includes first and second portions. The first portion includes information relating to a number of duplication times the unique user data corresponding to the user data entry is duplicated in the virtual memory space, and the second portion includes a selected part of the unique user data from which the unique user data may be reconstructed. The first portion may include an index to an extended reference counter table or a special data pattern table if the number of duplication times of the unique user data is greater than or equal to a predetermined number.

    Adaptive matrix multiplication accelerator for machine learning and deep learning applications

    公开(公告)号:US11475102B2

    公开(公告)日:2022-10-18

    申请号:US16407064

    申请日:2019-05-08

    Abstract: An adaptive matrix multiplier. In some embodiments, the matrix multiplier includes a first multiplying unit a second multiplying unit, a memory load circuit, and an outer buffer circuit. The first multiplying unit includes a first inner buffer circuit and a second inner buffer circuit, and the second multiplying unit includes a first inner buffer circuit and a second inner buffer circuit. The memory load circuit is configured to load data from memory, in a single burst of a burst memory access mode, into the first inner buffer circuit of the first multiplying unit; and into the first inner buffer circuit of the second multiplying unit.

    Overflow region memory management

    公开(公告)号:US10268413B2

    公开(公告)日:2019-04-23

    申请号:US15473311

    申请日:2017-03-29

    Abstract: A memory module includes a host interface configured to provide an interface to a host computer; one or more memory devices; a deduplication engine configured to provide a virtual memory capacity of the memory module that is larger than a physical size of the one or more memory devices; a memory controller for controlling access to the one or more memory devices; a volatile memory comprising a hash table, an overflow memory region, and a credit unit, wherein the overflow memory region stores user data when a hash collision occurs or the hash table is full, and wherein the credit unit stores an address of an invalidated entry in the overflow memory region; and a control logic is configured to control the overflow memory region and the credit unit and generate a warning indicating a status of the overflow memory region and the credit unit.

    METHOD AND APPARATUS FOR ENABLING LARGER MEMORY CAPACITY THAN PHYSICAL MEMORY SIZE

    公开(公告)号:US20170286313A1

    公开(公告)日:2017-10-05

    申请号:US15476757

    申请日:2017-03-31

    Abstract: A method of retrieving data stored in a memory associated with a dedupe module is provided. The method includes: identifying a logical address of the data; identifying a physical line ID of the data in accordance with the logical address by looking up at least a portion of the logical address in a translation table; locating a respective physical line, the respective physical line corresponding to the PLID; and retrieving the data from the respective physical line, the retrieving including copying a respective hash cylinder to the read cache, the respective hash cylinder including: a respective hash bucket, the respective hash bucket including the respective physical line; and a respective reference counter bucket, the respective reference counter bucket including a respective reference counter associated with the respective physical line.

Patent Agency Ranking