公开(公告)号：US20220404990A1

公开(公告)日：2022-12-22

申请号：US17638196

申请日：2020-06-23

Applicant: SUN YAT-SEN UNIVERSITY

Inventor： Zhiguang CHEN ,  Yutong LU

IPC: G06F3/06 ,  G06F16/13

Abstract: A low-latency file system file address space management method and system, and a medium. The method of the present invention comprises: generating a superblock and a block group allocation table from an address space of a storage device, wherein the superblock stores file system information and the allocation situation of block groups on a liner address space of the storage device, and the block group allocation table is used for marking the allocation situation of data blocks in the corresponding block group; when a file is created, dynamically creating or selecting a corresponding block group according to the size of a specified data block and allocating the data block; and writing file data into the allocated data block, and updating the block group allocation table and information of the superblock.

公开(公告)号：US20240028560A1

公开(公告)日：2024-01-25

申请号：US18039967

申请日：2021-04-25

Applicant: SUN YAT-SEN UNIVERSITY

Inventor： Zhiguang CHEN ,  Xianqi ZHENG ,  Yutong LU ,  Nong XIAO ,  Zejie HU ,  Jiawen LUO

IPC: G06F16/13 ,  G06F16/11 ,  G06F16/14

CPC classification number: G06F16/137 ,  G06F16/119 ,  G06F16/152

Abstract: A directory management method and system for a file system based on Cuckoo hash are provided, including the steps of reading metadata of a sub-directory or a sub-file, used as a target file, in a directory: receiving a request for reading the target file in the directory; for the target file, determining an ith candidate data block, in a hash table of the directory according to a hash calculation result obtained by performing hash calculation on a filename filename of the target file according to an ith hash function of Cuckoo hash iteratively; if the filename filename of the target file exists in the ith candidate data block, reading metadata of the target file and returning the metadata, and ending the process; otherwise, continuing to iterate until the iteration ends and returning a message that the target file does not exist.

公开(公告)号：US20240006026A1

公开(公告)日：2024-01-04

申请号：US18454977

申请日：2023-08-24

Applicant: SUN YAT-SEN UNIVERSITY

Inventor： Yutong LU ,  Ying WANG ,  Zhiguang CHEN

IPC: G16B30/20 ,  G16B40/30

CPC classification number: G16B30/20 ,  G16B40/30

Abstract: Disclosed are a genome assembly method, a genome assembly apparatus, a device and a storage medium. The method includes: obtaining a gene short sequence, and determining a first segmentation value; segmenting the gene short sequence based on the first segmentation value to obtain each gene subsequence; globally sorting each gene subsequence based on a preset grouped parallel sorting by regular sampling to obtain each sorted gene subsequence; traversing the distributed gene map in parallel to obtain each continuous gene sequence, and filling and assembling each continuous gene sequence to obtain each target continuous gene sequence; and determining a second segmentation value, and in response to that the second segmentation value is greater than or equal to a preset maximum segmentation threshold, assembling each target continuous gene sequence to obtain a genome assembly result.

公开(公告)号：US20230169005A1

公开(公告)日：2023-06-01

申请号：US17916033

申请日：2020-06-23

Applicant: SUN YAT-SEN UNIVERSITY

Inventor： Yutong LU ,  Zhiguang CHEN ,  Jia MA

IPC: G06F12/0862 ,  G06F12/02 ,  G06F12/0804

CPC classification number: G06F12/0862 ,  G06F12/0246 ,  G06F12/0804

Abstract: Provided in the present invention are a K-Truss graph-based storage system cache prefetching method, a system, and a medium, where method steps of the present invention include: when a data request stream in a system arrives, an access mode of the data request stream is determined; if said mode is a sequence mode, then n sequence blocks after a data block corresponding to the data request stream are prefetched and serve as prefetch data; otherwise, the data block corresponding to the data request stream serves as a query vertex to query a K-Truss graph, a truss structure matching the query vertex is obtained, and data of a data block from within the truss structure is obtained and serves as prefetch data; the K-Truss graph is updated according to the data block corresponding to the data request stream; and lastly a prefetch data block is prefetched into a main memory.

公开(公告)号：US20230161620A1

公开(公告)日：2023-05-25

申请号：US17917246

申请日：2020-06-23

Applicant: SUN YAT-SEN UNIVERSITY

Inventor： Yutong LU ,  Zhiguang CHEN

IPC: G06F9/48 ,  G06F9/50

CPC classification number: G06F9/4881 ,  G06F9/505

Abstract: Provided in the present invention are a Pull mode and Push mode combined resource management and job scheduling method and system, and a medium. The present invention includes parsing or identifying jobs to determine types, scheduling the jobs according to the types, and for high-performance computing jobs, distributing the jobs by using a Push mode: distributing computing nodes for the high-performance computing jobs, and pushing the high-performance computing jobs to the distributed computing nodes for execution; and for a big data processing job, using a Pull mode to perform distribution: waiting for a job request of each computing node, and when the computing node actively transmits the job request, transmitting the big data processing job to the computing node, which actively transmits the job request, for execution.

公开(公告)号：US20220027326A1

公开(公告)日：2022-01-27

申请号：US17279606

申请日：2019-03-29

Applicant: SUN YAT-SEN UNIVERSITY

Inventor： Yutong LU ,  Zhiguang CHEN ,  Nong XIAO ,  Fang LIU

IPC: G06F16/185 ,  G06F16/13

Abstract: A metadata management method, system and medium are provided. All directories and files in a file system are organized in a tree structure, and all directories and subfiles under a single directory are saved as a whole on a continuous address space of an underlying storage device in the order of creation in sequence. When metadata is updated, update operations on all subdirectories/subfiles under a directory are appended to the tail of the directory in the form of log in order, so that each metadata operation triggers only one IO request. In the process of metadata retrieval, parsing a subdirectory under a parent director requires reading all data on the continuous address space corresponding to the parent directory, and then searching for a match by means of multiple acceleration methods.