US 11,868,315 B2
Method for splitting region in distributed database, region node, and system
Jieshan Bi, Shenzhen (CN); Hao Fang, Shenzhen (CN); Chaoqiang Zhong, Shenzhen (CN); and Yijun Guo, Shenzhen (CN)
Assigned to Huawei Cloud Computing Technologies Co., Ltd., Gui'an New District (CN)
Filed by Huawei Cloud Computing Technologies Co., Ltd., Gui'an New District (CN)
Filed on Apr. 23, 2018, as Appl. No. 15/960,167.
Application 15/960,167 is a continuation of application No. PCT/CN2016/080144, filed on Apr. 25, 2016.
Claims priority of application No. 201510767508.9 (CN), filed on Nov. 11, 2015.
Prior Publication US 2018/0239780 A1, Aug. 23, 2018
Int. Cl. G06F 16/17 (2019.01); G06F 16/23 (2019.01); G06F 16/901 (2019.01); G06F 16/28 (2019.01); G06F 3/06 (2006.01)
CPC G06F 16/173 (2019.01) [G06F 3/061 (2013.01); G06F 16/23 (2019.01); G06F 16/28 (2019.01); G06F 16/901 (2019.01)] 19 Claims
OG exemplary drawing
 
1. A method, comprising:
reserving, by a region node, a first region of a database, wherein the first region is blank and is not used for current data storage, the first region is reserved during current data storage for storing data of a first key-value type of a plurality of key-value types, and the first region is a first key-value range between a first key-value and a second key-value, and wherein reserving the first region of the database comprises:
pre-estimating, according to a type of currently stored data, a second key-value range, and reserving the first region to have the first key-value range according to the pre- estimating to determine that that the currently stored data does not fall into the blank first region, wherein the type of currently stored data is the first key-value type of the plurality of key-value types;
receiving, by the region node, a region split command when first data of the first key-value type is to be stored in the first region of the database, wherein the region split command comprises split point information of the first region, and the region split point information indicates at least one point located inside the blank first region to use to split the first region;
splitting, by the region node, the first region into at least two second regions according to the split point information;
after splitting the first region into the at least two second regions, storing, by the region node, the first data of the first key-value type in the at least two second regions; and
instructing, by the region node, to record, in a metadata table, location information of the region node where the at least two second regions are located, so that the region node where the second regions are located are findable according to the metadata table.