Host Multipath Layer Notification for iSCSI Path Switchover
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current computational environments face inefficiencies in detecting failed nodes and redirecting I/O operations, leading to poor performance and resource wastage due to reliance on I/O timeouts and resource-intensive keepalive functions.
Innovation Solution
A computer-implemented method using asynchronous messages to quickly notify hosts of node failures and redirect I/O operations to failover paths, including an indication of failed paths and alternative IP addresses, allowing for prompt failover and failback operations before and after an expiration time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If I/O timeout methods are used to detect node failures, then the system can detect failures, but the detection time is delayed and I/O operations are wasted on failed nodes
Solution Approach 1:
The system performs preliminary actions by sending asynchronous messages from the storage system to the host upon detecting node failures, before the host's I/O timeout mechanisms would naturally detect the failure. This proactive notification allows the host to switch to failover paths immediately, avoiding the time loss associated with waiting for timeout-based detection.
2Reliability
If keepalive functions are used to monitor node status, then node failures can be detected, but the resource consumption is high
Solution Approach 1:
The storage system performs self-service by autonomously detecting its own node failures and generating asynchronous failure notification messages. This eliminates the need for separate keepalive monitoring functions on the host side, reducing resource consumption while maintaining reliable failure detection through the storage system's internal monitoring capabilities.
3Ease of operation
If the host continues I/O operations on failed paths until timeout, then simple operation is maintained, but performance deteriorates due to wasted I/O operations
Solution Approach 1:
The system implements feedback by sending asynchronous failure notification messages from the storage system to the host, providing real-time information about failed paths. This feedback mechanism allows the host to immediately adjust its I/O operations and switch to failover paths, maintaining simple operation while preventing performance deterioration from continued I/O on failed paths.
4Device complexity
If failover paths are not pre-configured with alternative IP addresses, then system complexity is reduced, but failover capability is compromised
Solution Approach 1:
The system performs preliminary action by pre-configuring failover paths with alternative IP addresses and storing them in the asynchronous failure notification messages. This preparation is done in advance, so when a failure occurs, the host can immediately use the pre-configured failover paths without complex runtime decision-making, maintaining both simplicity and reliability.
Data Source
AI summary
Described herein are techniques for managing failover in a data center environment interconnected using an internet small computer systems interface (iSCSI) communication protocol, the techniques including receiving, at a host and from a kernel driver, an asynchronous message comprising an indication of a failed path associated with a first node having a first port, a list of internet protocol (IP) addresses associated with a plurality of failover paths including a first failover path associated with a second node having a second port, and an expiration. The techniques further including performing, by the host and before the expiration, a first input/output (I/O) operation on the second port associated with the second node. The techniques further including performing, by the host and after the expiration, a second I/O operation on the first port of the first node.