Low-energy storage system and method for cloud video surveillance data based on sla classification

Abstract

The invention discloses a low-energy storage system and method for cloud video monitoring data based on SLA classification. By using an SLA with an access time period, virtual machines running monitoring and browsing tasks and data storage nodes in the cloud video monitoring system Classify according to the access time period, so as to divide the whole day cloud video service into different time periods for classification processing; on this basis, design a storage node operation schedule to monitor the operation status of all storage nodes in the data center Control, so that the storage nodes not in the access time period are in the energy-saving state, while the storage nodes in the access time period are in the normal operation state, so as to save the power consumption of the storage nodes under the premise of ensuring the SLA.

Description

technical field [0001] The invention relates to a low-energy storage system and method for cloud video surveillance data based on SLA classification, belonging to the field of cloud video surveillance. Background technique [0002] Cloud computing is a computing model that uses the Internet to access shared resource pools (such as computing facilities, storage devices, applications, etc.) anytime, anywhere, on demand, and conveniently. A new cloud computing service model of "Video Surveillance as a Service". Cloud video surveillance services have greatly reduced the construction and maintenance costs of users. According to the statistics of IMS Research, the demand for video surveillance services based on cloud computing is growing at a rate of 20% to 30% per year. Therefore, the video monitoring system based on cloud computing is the future development trend in the monitoring field. The cloud video surveillance data center is the carrier of all surveillance video resource...

AI Technical Summary

Problems solved by technology

[0004] On the premise of providing a certain fault tolerance, the method of static data placement makes certain nodes not provide data access and storage services for a certain period of time, so as to shut down or suspend these nodes to achieve energy saving; however, the method of static data placement does not consider cloud monitoring Due to the large storage space of video data and the strong randomness of user access, it is difficult to directly apply it to the cloud video surveillance system

The dynamic placement storage management method dynamically adjusts the location of data storage according to the data access mode or frequency, and migrates the data with high access frequency to the cache or some nodes, so that other nodes have no access requests for a long period of time, and make It enters a low energy consumption state to save energy; however, cloud video surveillance is a data-intensive application, and the dynamic placement storage management method will occupy a large amount of network bandwidth during the video data migration process, and affect system service performance. The migration cost is too high, so it is not suitable for Streaming media storage, so it is difficult to apply to cloud video surveillance occasions

Images