Computing power load scheduling method and device for distributed video intelligent analysis platform
A distributed video and intelligent analysis technology, which is applied in the directions of multi-programming devices, computing, resource allocation, etc., to achieve the effects of ensuring reliability, fast response, and improving user experience
Pending Publication Date: 2021-06-04
招商新智科技有限公司
View PDF0 Cites 4 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0005] In view of this, the purpose of one or more embodiments of this specification is to propose a computing power load scheduling method and device for a distributed intelligent video analysis platform, so as to solve the problem of load balancing scheduling of the computing power of the intelligent video analysis platform
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment approach
[0050] As an implementation manner, step 103 includes:
[0051] 1. Determine whether each algorithm service satisfies the load capacity invoked by the user, if so, continue to obtain the optimal algorithm service, and if not, continue to traverse the next algorithm service;
[0052] The algorithm resource consumption information includes GPU video memory consumption estimation GM, memory capacity consumption estimation M, and CPU occupancy estimation C0 corresponding to each algorithm type of each manufacturer, and the computing power resource consumption information includes the algorithm type corresponding to each algorithm service Support list TList, manufacturer support type, CPU occupancy rate C1, GPU memory free capacity GMF and memory free capacity MF, the user request data includes the user request algorithm type T0;
[0053] Judging whether the user request algorithm type T0 is in the algorithm type support list Tlist, if not included, then return FALSE;
[0054] If ...
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
Login to View More Abstract
One or more embodiments of the invention provide a computing power load scheduling method and device for a distributed video intelligent analysis platform, and due to the adoption of a real-time computing power resource acquisition method, a load scheduling calculation result can be more real-time and more accurate; a method of combining algorithm resource consumption estimation and algorithm service real-time resource idle rate is adopted in a computing power load scheduling calculation process, so that a result of a load scheduling algorithm has algorithm service computing power estimation capability, and the reliability of the scheduling result is ensured; due to the fact that calculation and comparison of the load priority value are adopted in the computing power load scheduling calculation process, selection of the optimal load algorithm service of the system is achieved, the user algorithm calling calculation response speed is higher, and the user experience is improved.
Description
technical field [0001] One or more embodiments of this specification relate to the technical field of computing power load scheduling, and in particular to a computing power load scheduling method and device for a distributed intelligent video analysis platform. Background technique [0002] In recent years, with the rapid development of machine learning and deep learning, more and more products have become more automated and intelligent through the application of image analysis technology, such as face recognition and license plate recognition. The increasing number of client-side calls has overwhelmed many video intelligent analysis platforms, forcing many platform manufacturers to respond by improving the performance of hardware facilities or increasing the number of hardware servers. However, this response method not only increases the cost of the platform, but also increases the reduce the workload of platform maintenance. Therefore, researching a sufficient and flexib...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06F9/50
CPCG06F9/5083G06F9/5016G06F9/505
Inventor 陈卫强杜渐段洪琳张威奕张凯王兆明王库超
Owner 招商新智科技有限公司