Beijing Oceanbase Technology Co., Ltd.

Alibaba Group Holding Limited

Ant Financial (hang Zhou) Network Technology Co., Ltd.

Advanced New Technologies Co., Ltd.

Yuzhong Zhao

Hu Xu

Zhenkun Yang

Chenxi Li

Wenhui Shi

Zhenjiang Xie

Xiao Chen

Hui Cao

Min Yang

Jianyun Sun

Runchen Hu

The specification provides an IO request scheduling method. The method includes: obtaining a target IO request to be scheduled, and determining whether the target IO request is a traffic-limited IO request: when the target IO request is a traffic-unlimited first-type IO request, determining a current moment as a scheduling moment of the target IO request: or when the target IO request is a traffic-limited second-type IO request, determining a scheduling moment of the target IO request based on a current resource traffic-limiting proportion; and adding the target IO request to an IO request set, where several IO requests in the IO request set are sequentially sent to a disk for processing in ascending order of scheduling moments.

IO request scheduling methods and apparatuses

Implementations of the present specification describe evicting a memory block in a cache, including determining a temperature index value of a memory block in a cache, where the memory block is a storage unit of the cache. Based on the temperature index value of the memory block in the cache, determining a to-be-evicted memory block that needs to be cleared in the cache. A target element in the to-be-evicted memory block is migrated to a target memory block in the cache, where the target memory block is a memory block that is retained in the cache and whose available storage space is greater than storage space required by the target element. The to-be-evicted memory block is deleted from the cache.

Methods, apparatuses, and electronic devices for evicting memory block in cache

A data compression method includes: obtaining a to-be-compressed object; searching a recommendation record for a recommended compression coding rule that meets a compression rate condition, the recommendation record being configured to record a compression coding rule of a historical compressed object and corresponding compression rate information, and the historical compressed object being of a same type as the to-be-compressed object; and if the recommended compression coding rule that meets the compression rate condition is found, compressing the to-be-compressed object by using the recommended compression coding rule; and if the recommended compression coding rule that meets the compression rate condition is not found, starting a regular compression coding process to obtain estimated compression rates of a plurality of compression coding rules for the to-be-compressed object, selecting a target compression coding rule based on at least the estimated compression rates, and compressing the to-be-compressed object by using the target compression coding rule.

Data compression method and apparatus, and computer device

The present specification discloses a database migration method, including: copying metadata from a first server of a database, the metadata including identification information of synthetic data; receiving a service instruction sent by a client after the copying the metadata from the first server of the database; determining, based on the service instruction, whether the synthetic data needs to be retrieved from the first server of the database; and in response to it is determined that the synthetic data needs to be retrieved from the first server of the database, copying the synthetic data from the first server of the database based on the identification information of the synthetic data. According to the present disclosure, the service instruction sent by the client is received in advance, and the synthetic data is copied from the first server of the database based on the identification information of the synthetic data, thereby reducing time taken for database migration.

Database migration method, apparatus, device, and computer-readable medium

Implementations of the present application provide a partition splitting method. Data files in a target source partition are divided into at least two data groups. A first data group from the at least two data groups is selected as a target splitting data group. A first splitting step is performed on the target splitting data group, including: splitting the target splitting data group into a plurality of data sub-groups based on a predetermined splitting rule, wherein each data sub-group of the plurality of data sub-groups corresponds to a new partition of at least two new partitions for the target source partition, and assigning each data sub-group to the new partition corresponding to the data sub-group. A second data group from the at least two data groups is selected as the target splitting data group, wherein the second data group is different from the first data group.

Partition splitting method and apparatus, electronic device, and readable storage medium

One or more implementations of the present specification disclose a database analysis information generation method, including: reading data in a to-be-merged data level and data in a merging data level; generating a new merging data level by merging the data in the to-be-merged data level and the data in the merging data level in a merging process; and performing an operation of generating new database analysis information based on data in the new merging data level and data in existing database analysis information in the process of generating the new merging data level.

Database management method and system, and electronic device

This disclosure relates to bitmap index writing, dumping, merging, and querying. In one aspect, a method includes determining, based on update data, a to-be-updated row of a bitmap index that indexes data of a database using a bitmap data structure. The to-be-updated row of the bitmap index is locked. The locking includes adding a row lock to the to-be-updated row, thereby preventing the to-be-updated row from being updated by other update operations. At least one other row of the bitmap index remains unlocked for updating by other update operations. Data in the to-be-updated row of the bitmap index is updated based on the update data.

Updating and querying a bitmap index

Embodiments provide a data block storage method applied to a database employing an LSM tree. According to the method, data in the 0layer of the LSM tree is organized in MacroBlocks. When the data in the LSM tree is transferred from an upper layer to a lower layer, the data is transferred to the lower layer in the form of MacroBlocks, so that all layers of the LSM tree are organized in the form of MacroBlocks. When inter-layer data merger is performed, the merger may be performed based on the MacroBlocks, which can avoid rewriting of all data on an entire layer and can effectively reduce disk IO and temporary space consumption. Moreover, as each MacroBlock comprises a plurality of variable-length MicroBlocks, the data may be stored in a storage space in a more compact manner.

Data block storage method and apparatus, and electronic device

One example method includes receiving, by the one or more processors, a request to update data stored on a distributed data storage system that comprises a primary server and a plurality of secondary servers, the request comprising data updates; sending, by the one or more processors, an update log instruction to the primary server and the plurality of secondary servers so that each of the primary server and the plurality of secondary servers records a log associated with the request; determining, by the one or more processors, that the log has been recorded; and in response to determining that the log has been recorded, sending, by the one or more processors, a data storage instruction to the primary server and a secondary server of the plurality of secondary servers so that each of the primary server and the secondary server stores the data updates.

Growing community of inventors

Hangzhou, China

Yuzhong Zhao