Network Storage Performance Monitoring via Traffic Mirroring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In networked storage environments, it is challenging to efficiently identify and manage performance bottlenecks in storage volumes, as existing solutions require installing agents on client machines, which is inefficient and complex, especially in large-scale environments.
Innovation Solution
A remote acquisition unit (RAU) is used to mirror network traffic at network switches, parsing multiple protocol layers to determine impacted storage volumes and client access addresses, providing this information to a management console for simplified performance data and corrective action.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If agents are installed on client machines to collect traffic information, then performance monitoring capability is improved, but device complexity and ease of operation deteriorate due to software compatibility and maintenance requirements
Solution Approach 1:
The patent introduces network switches as intermediary devices that collect and analyze traffic information between clients and storage systems. The switches mirror traffic to a management console, eliminating the need for agents on client machines. This intermediary approach maintains measurement precision while reducing device complexity and maintenance burden.
Solution Approach 2:
The patent uses traffic mirroring functionality where network switches create copies of traffic flows and send them to a management console for analysis. This copying mechanism allows performance monitoring without installing agents on client machines, resolving the contradiction between monitoring capability and system complexity.
2Loss of information
If agents are deployed on thousands of client machines, then traffic information collection is improved, but productivity and ease of operation worsen due to continuous maintenance efforts
Solution Approach 1:
Network switches serve as intermediaries that automatically collect and mirror traffic information to the management console. This eliminates the need to deploy and maintain agents on thousands of client machines, significantly reducing productivity loss while maintaining complete traffic information collection.
Solution Approach 2:
The network switches perform self-service by automatically mirroring traffic flows to the management console without requiring external agent software. This self-service capability eliminates continuous maintenance efforts while ensuring complete traffic information is captured for analysis.
3Device complexity
If storage array operates blindly without client system information, then system simplicity is maintained, but measurement precision deteriorates making it difficult to identify performance bottleneck sources
Solution Approach 1:
The network switch acts as an intermediary that captures detailed traffic information between clients and the storage array. The management console analyzes this mirrored traffic to identify performance bottlenecks and map them to specific clients and storage volumes, providing measurement precision without modifying the storage array's simplicity.
Solution Approach 2:
The system copies traffic information from the network path between clients and storage array to a management console for analysis. This copying approach enables precise performance bottleneck identification while keeping the storage array configuration simple and unchanged.
Data Source
AI summary
Methods and systems for a networked storage environment are provided. One method includes mirroring a plurality of requests from a switch and transmitting the mirrored plurality of requests to a remote acquisition unit; extracting application layer protocol data units from assembled transport layer packets; parsing the application layer protocol data units to obtain file system requests; identifying storage volume identifiers from the parsed file system requests that are associated with a greatest number of operations; identifying network addresses for client systems initiating the greatest number of operations for the storage volumes and network addresses of target storage systems managing the storage volumes; and providing a total number of operations for the plurality of requests in a given time, the identified storage volume identifiers, the network addresses of the client systems and the network addresses of the target storage systems to a management console.