Method and device for processing commands based on iSCSI (internet small computer system interface)

Abstract

The invention discloses a method and device for processing commands based on an iSCSI (internet small computer system interface). The method comprises the following steps: after a memory system establishes the iSCSI session with a host machine, receiving service parameters which are set for each LUN (logic unit number) in the iSCSI session from the host machine, wherein the service parameters comprise maximum response time; adopting the memory system to receive an iSCSI IO (input output) command sent from the host machine, wherein service optimization markers are configured in the command and comprise critical data markers and / or response time limitation markers; using the memory system to judge whether the response time limitation markers are configured in the command; if the response time limitation markers are configured in the command, judging whether the critical data markers are configured in the command; if the critical data markers are configured in the command, preferentially scheduling the command; if the critical data markers are not configured in the command, calculating the finish time of the command; and calculating the difference between the finish time of the command and the existing time, if the difference is less than a preset threshold value, preferentially scheduling the command; and otherwise normally scheduling the command. According to the invention, the processing efficiency of the command based on the iSCSI can be improved.

Description

technical field [0001] The invention relates to the field of storage technology, in particular to a command processing method and device based on an Internet Small Computer System Interface (iSCSI, Internet Small Computer System Interface). Background technique [0002] At present, the Internet Protocol Storage Area Network (IPSAN, Internet Protocol Storage Area Network) is widely used as the storage body in the monitoring system, and the data of the camera is stored in the IPSAN. When the user needs to view the historical video, the video data is read from the IPSAN. . [0003] figure 1 This is a typical existing IP monitoring network diagram. Such as figure 1 As shown, it mainly includes: [0004] Video storage process: the encoder (EC, Encoder) connects to IPSAN through iSCSI, and writes the video data into IPSAN. [0005] Video playback process: the video client (VC, Video Client) or decoder (DC, Decoder) sends an application to the video management (VM, Video Manag...

AI Technical Summary

Problems solved by technology

[0006] The video storage process has relatively high requirements for real-time storage. Due to the limited EC cache resources, when the array on the IPSAN is being rebuilt or synchronized, the response to the input and output (IO, Input Output) of the client will slow down. The large size causes the EC cache data to be flushed very slowly, which eventually leads to EC storage packet loss

[0007] For the video playback process, when there is multi-channel playback, since the DM and IPSAN only establish one iSCSI session, all playback services run on one session, and each session has only one command queue, causing multiple playback services to interact with each other. Impact, when the IO response of the playback service on one of the storage logical units is slow, the IO response of the playback service on other storage logical units will also be slow

Images