	US 12,493,422 B2
	Resource allocation method and apparatus
Zhou Jiang, Chengdu (CN); Zhuo Cheng, Zürich (CH); Haitao Shi, Chengdu (CN); Yuanyuan Wang, Chengdu (CN); and Weijie Xiang, Hangzhou (CN)
Assigned to HUAWEI TECHNOLOGIES CO., LTD., Shenzhen (CN)
Filed by Huawei Technologies Co., Ltd., Shenzhen (CN)
Filed on Jan. 22, 2024, as Appl. No. 18/419,111.
Application 18/419,111 is a continuation of application No. PCT/CN2022/082051, filed on Mar. 21, 2022.
Claims priority of application No. 202110838151.4 (CN), filed on Jul. 23, 2021.
Prior Publication US 2024/0201880 A1, Jun. 20, 2024
Int. Cl. G06F 3/06 (2006.01)

CPC G06F 3/0631 (2013.01) [G06F 3/0604 (2013.01); G06F 3/0644 (2013.01); G06F 3/067 (2013.01)]

20 Claims

1. A method implemented by one or more processors, comprising:

outputting first models of first cache objects based on status informations of the first cache objects, wherein the status informations comprise historical information and at least one of a reuse distance information or a memory eviction information, wherein the historical information comprises a correspondence that is between a page resource allocated to each of the first cache objects and a first data read hit ratio of each of the first cache objects and that exists in an execution process of a historical service, wherein the reuse distance information indicates a quantity of other data blocks between two consecutive accesses to a same data block in the first cache objects, wherein the memory eviction information indicates changes in hit ratios of the first cache objects under a condition of evicting partial page resources, and wherein the first models indicate sizes of first page resources to be allocated to the first cache objects respectively when the first cache objects meet various hit ratios;

outputting a second model based on the first models, wherein the second model indicates a target page resource to be allocated to each of the first cache objects; and

adjusting the first page resources based on the second model.