Unlock instant, AI-driven research and patent intelligence for your innovation.

A hybrid multi-mode hot data caching strategy based on depth learning

A caching strategy and deep learning technology, applied in digital data processing, memory systems, instruments, etc., can solve the problems of reducing hit rate and Cache work efficiency, and achieve the goal of improving index query performance, effective memory space, and reducing access times Effect

Inactive Publication Date: 2019-03-29
CHINA UNIV OF PETROLEUM (EAST CHINA)
View PDF5 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage is that the hit rate and Cache work efficiency are reduced

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A hybrid multi-mode hot data caching strategy based on depth learning
  • A hybrid multi-mode hot data caching strategy based on depth learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative work shall fall within the protection scope of the present invention.

[0025] Such as figure 1 As shown, the process of the intelligent prediction method based on the deep learning DBN algorithm mainly includes three parts: preprocessing of data, DBN pretraining, optimization of BP algorithm, and post-processing of prediction results.

[0026] Combine below figure 1 versus figure 2 , Explain in detail the specific process of the intelligent prediction method based on deep learning:

[0027] Step (1), aggregat...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a hybrid multi-mode hot data caching strategy based on depth learning. The design goal of hierarchical indexing model is to quickly respond to most of the query requests in theindex cache layer, reduce the access times of the persistent storage layer and improve the overall performance of the indexing system. But the capacity of cache is limited after all, how to choose anappropriate cache replacement strategy, in the case of using as few caches as possible, to improve the hit rate of the index system is an important research issue, so we propose a hybrid multi-mode hot data cache replacement strategy based on deep learning. Based on the original multi-mode hot data sensitive caching strategy, By supporting the query trend prediction service provided by the platform, The query task is predicted, and the DBN prediction algorithm in depth learning is used to predict the next possible query task after the completion of the query task, and the index records relatedto the predicted query task are loaded into the cache layer, so as to increase the hit rate of the index records in the cache and improve the query efficiency.

Description

Technical field [0001] The invention relates to a cache replacement strategy, intelligent prediction and deep learning, in particular to an intelligent prediction method based on deep learning. Background technique [0002] By quickly responding to most query requests at the index cache layer, the number of accesses to the persistent storage layer is reduced, thereby improving the overall performance of the indexing system. Since the capacity of the cache is limited, it is necessary to select an appropriate cache replacement strategy to increase the hit rate and improve the overall performance of the indexing system when using as little cache as possible. In the memory index service process, frequently accessed data can be stored in the index cache layer, thereby more effectively using memory space and improving index query performance. The technologies closest to the present invention are: [0003] (1), Least Frequently Used (LFU) algorithm: LFU (Least Frequently Used) algorithm...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F12/0895
CPCG06F12/0895
Inventor 张卫山房凯任鹏程
Owner CHINA UNIV OF PETROLEUM (EAST CHINA)