Title:
RESOURCE MANAGEMENT METHOD AND DEVICE FOR DISTRIBUTED MACHINE LEARNING TASKS
Document Type and Number:
WIPO Patent Application WO/2022/120979
Kind Code:
A1
Abstract:
The present invention relates to the field of machine learning tasks, and specifically relates to a resource management method and device for distributed machine learning tasks. The method and device comprise a user submitting a machine learning task, and the task comprises two aspects of information, the first being the size of a dataset, and the second being the number of containers; a prediction model calculates the allocation size of a memory according to the size of the dataset and the number of containers while selecting a corresponding cache mode; according to the selection of the cache mode, the allocation of the memory can be divided into two circumstances: an optimal performance model is selected when the memory is sufficient; and an optimal resource utilization model is selected when the memory is insufficient. The present invention mainly analyzes the attributes of distributed machine learning and calculates the resource management for a frame, and according to the foregoing analysis, constructs a model for memory prediction and cache mode selection, which directly allocates memory and selects a cache mode for a new machine learning task without the need for additional application portraits.
More Like This:
JP2023089786 | CONTROL PROGRAM, METHOD FOR CONTROL, AND DISPERSION FILE SYSTEM |
JP5266197 | Data center task mapping |
WO/2020/026456 | SERVER DEVICE AND SYSTEM |
Inventors:
LUO SHUTIAN (CN)
YE KEJIANG (CN)
XU CHENGZHONG (CN)
YE KEJIANG (CN)
XU CHENGZHONG (CN)
Application Number:
PCT/CN2020/139264
Publication Date:
June 16, 2022
Filing Date:
December 25, 2020
Export Citation:
Assignee:
SHENZHEN INST OF ADV TECH CAS (CN)
International Classes:
G06F9/50
Domestic Patent References:
WO2014188052A1 | 2014-11-27 |
Foreign References:
CN110427263A | 2019-11-08 | |||
CN105824737A | 2016-08-03 | |||
CN109961151A | 2019-07-02 |
Other References:
ZHANG SUFANG; ZHAI JUNHAI, WANG CONG, SHEN CHU: "Big Data and Big Data Machine Learning", JOURNAL OF HEBEI UNIVERSITY (NATURAL SCIENCE EDITION), vol. 38, no. 3, 25 May 2018 (2018-05-25), pages 299 - 308, 336, XP009537475, ISSN: 1000-5854
Attorney, Agent or Firm:
BEIJING ZHONG XUN TONG DA INTELLECTUAL PROPERTY AGENCY CO., LTD. (CN)
Download PDF:
Previous Patent: POP-UP SYNCHRONOUS REFINING AND FILTERING DEVICE FOR SOLID WASTE
Next Patent: AIR CONDITIONER INDOOR UNIT AND AIR CONDITIONER
Next Patent: AIR CONDITIONER INDOOR UNIT AND AIR CONDITIONER