Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Big data tensor canonical decomposition calculation method based on Shenwei multi-core processor

A technology of many-core processor and calculation method, applied in the field of canonical decomposition calculation of big data tensor based on Shenwei many-core processor, which can solve problems such as acceleration

Active Publication Date: 2019-10-22
BEIHANG UNIV
View PDF6 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The technology of the invention solves the problem: it reduces the difficulty for programmers to write parallel programs on Shenwei, does not require programmers to learn the programming method of Shenwei architecture, and provides a big data tensor canonical decomposition calculation based on Shenwei many-core processors method, realizes the tensor canonical decomposition calculation algorithm swTensor combined with MapReduce, accelerates the tensor canonical decomposition process, improves the efficiency of tensor canonical decomposition, and thus improves Shenwei's computing power

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
  • Big data tensor canonical decomposition calculation method based on Shenwei multi-core processor
  • Big data tensor canonical decomposition calculation method based on Shenwei multi-core processor
  • Big data tensor canonical decomposition calculation method based on Shenwei multi-core processor

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0055] In order to make the purpose, technical solution and advantages of the present invention more clear, the present invention will be further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific examples described here are only used to explain the present invention, not to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not constitute a conflict with each other.

[0056] The system architecture diagram of the present invention is as follows figure 1 As shown, the execution flow diagram is as follows Figure 5 As shown, the entire job is started by the MPE, the task scheduler is executed by the CPE, the Task is the original data stored in the main memory, and the Results are the partial results stored in the main memory after the CPE processing is completed...

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 relates to a big data tensor canonical decomposition calculation method based on a Shenwei multi-core processor. The big data tensor canonical decomposition calculation method comprisesthe following steps: 1) proposing a big data calculation method swMR according to characteristics of the Shenwei multi-core processor; 2) dividing computing processing unit cluster CPE organized in a8 * 8 grid form by the swMR in the Shenwei multi-core processor into 32 CPE pairs, grouping two adjacent CPEs in each row in the grid, processing a mapping Map task by one CPE, and processing a reduction Reduce task by the other CPE; 3) according to the workload condition of each CPE pair, mapping reduction Map / Reduce processing role dynamic conversion is carried out in the CPE pair to realize dynamic adjustment of the workload; and 4) based on swMR and an Shenwei multi-core processor, providing a reasonable calculation method swTensor for tensor canonical decomposition calculation. Implementation of a machine learning algorithm and tensor canonical decomposition calculation on an Shenwei processor is supported; dynamically dividing workloads to balance job distribution conditions in the CPE pair. Based on a mapping reduction MapReduce programming model, the swTensor efficiently supports tensor canonical decomposition calculation.

Description

technical field [0001] The invention relates to the fields of concurrent execution of many-core processors, tensor canonical decomposition calculation and MapReduce programming method, and in particular to a big data tensor canonical decomposition calculation method based on Shenwei many-core processor. Background technique [0002] The development of the Internet recommendation system has greatly improved the efficiency of users' online browsing, saved users' time, and helped users quickly find the products or information they need. The recommendation system stores and calculates feature information in the form of tensors. In addition, tensors also play an important role in the fields of computer vision, image processing, and signal processing. The application of tensors greatly facilitates the storage and representation of data such as feature information, which improves the efficiency of writing and running applications. Tensor canonical decomposition is an important tec...

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): G06F17/10G06F17/16
CPCG06F17/10G06F17/16
Inventor 杨海龙钟小刚栾钟治
Owner BEIHANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com