Optical fiber storage network big data intelligent monitoring and early warning method and system

Abstract

The invention relates to the field of storage area networks, and discloses an optical fiber storage network big data intelligent monitoring and early warning method and system, and the method comprises the steps: collecting bottom index information of an SAN optical fiber switch for collection; storing the underlying index information into an information database; identifying an abnormal value ofthe bottom-layer index information by utilizing an artificial intelligence engine component; and enabling the information display component to operate the information database in real time and displaythe information according to the bottom-layer index information abnormal value output by the artificial intelligence engine component. By establishing the artificial intelligence engine assembly, thestates and early warning information of the corresponding switches and ports are output through big data processing, the early warning accuracy is high, a client display interface is provided throughthe information display assembly, the problems and conditions of the switches can be conveniently monitored and analyzed, a user can obtain early warning of possible problems of the SAN network switch due to module instability, link instability and link congestion spreading in time, and the method and the system provide timely guidance for preventive maintenance.

Description

technical field [0001] The invention relates to the field of storage area networks, in particular to a method and system for intelligent monitoring and early warning of big data in an optical fiber storage network. Background technique [0002] At present, the centralized data storage of large-scale user data centers uses optical fiber storage switches as the core of the storage area network (Storage Area Network, SAN) connection, and the problems of storage switches and storage networks will directly affect data access, and then affect databases and core transaction systems. , there are potential risks like business interruption, transaction failure, database slowdown / downtime, disaster recovery failure, and even transaction data loss. Many of the reasons for these events are not traditional equipment component failures, but the business impact caused by the occurrence is greater than that of component failures, including: (1) port module aging. Module aging is an inevitabl...

AI Technical Summary

Problems solved by technology

[0002] At present, the centralized data storage of large-scale user data centers uses optical fiber storage switches as the core of the storage area network (Storage Area Network, SAN) connection, and the problems of storage switches and storage networks will directly affect data access, and then affect databases and core transaction systems. , there are potential risks like business interruption, transaction failure, database slowdown / downtime, disaster recovery failure, and even transaction data loss

Many of the reasons for these events are not traditional equipment component failures, but the business impact caused by the occurrence is greater than that of component failures, including: (1) port module aging. Module aging is an inevitable phenomenon of equipment. During the aging process, all indicators have not reached the fault level, and there will be no fault alarm, but the host-side path timeout and even the database timeout phenomenon caused by aging caused by congestion, verification errors, and code errors may still exist. It is difficult to pass the traditional monitoring and alarm, which has a significant impact on business continuity; (2) the slow device problem, in a complex SAN network, if there is a slow device, it may affect the local area, and then spread the problem to the entire switch and level Connected ports cause events affecting the entire network and services on the entire SAN network

[0003] However, the existing SAN monitoring and alarming products and methods cannot achieve timely detection and rapid warning of faults, and there are many defects

Some products provide fault alarms through SNMP, only power fan and board fault alarms, and lack preventive measures such as firmware aging and slow device impact; some products and methods even take into account the monitoring of sending and receiving light, bit errors, and blockages However, due to the low accuracy rate of a single indicator and the inability to determine the exact threshold, most of its monitoring alarms do not match the actual situation, and serious false positives and missed negatives cannot be used as a basis for user maintenance actions.

Images