Live Suspend Resume I/O Traffic Management
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In large computer systems, maintenance or replacement of I/O devices often requires shutting down parts of the system, leading to extended downtime and reduced business efficiency, as existing technologies lack efficient methods to suspend and resume I/O traffic without disrupting the entire system.
Innovation Solution
A method involving a Live Suspend and Resume (LSR) module that sends suspend and resume requests to I/O devices, managing their states to pause and resume I/O traffic types, allowing for maintenance without shutting down the entire system by queuing messages and restoring device settings.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of repair
If the computer system is shut down to maintain I/O devices, then the I/O devices can be maintained, reset, or replaced, but the system experiences extended downtime and reduced business efficiency
Solution Approach 1:
The patent segments the system into online and offline I/O device groups, allowing maintenance operations to be performed on offline devices while online devices continue to operate. The bus controller divides I/O bus bandwidth dynamically, allocating resources to maintenance operations without completely shutting down the system, thus resolving the contradiction between ease of repair and productivity.
Solution Approach 2:
The patent implements dynamic state transitions for I/O devices between online, offline, suspended, and resumed states. The bus controller dynamically adjusts I/O bus bandwidth allocation based on device state and maintenance requirements, enabling flexible resource management that maintains productivity while facilitating repairs.
2Ease of repair
If the entire system is shut down for I/O device maintenance, then all I/O devices can be accessed and maintained, but the downtime is extended and system availability is reduced
Solution Approach 1:
The patent segments I/O devices into maintainable groups that can be independently suspended and resumed. By dividing the system into operational and maintenance zones, devices requiring maintenance can be isolated and serviced without affecting other devices, significantly reducing the time loss compared to complete system shutdowns.
Solution Approach 2:
The patent performs preliminary suspension of I/O devices before maintenance operations begin, preserving their state information. This allows devices to be quickly resumed after maintenance without requiring full system reboot, thereby reducing downtime and loss of time.
3Reliability
If I/O traffic is completely stopped for device maintenance, then device state can be preserved and maintenance can be performed, but system productivity is reduced during the maintenance period
Solution Approach 1:
The patent dynamically controls I/O traffic by suspending it selectively to specific devices rather than stopping all traffic. The bus controller manages traffic flow dynamically, redirecting I/O requests away from devices under maintenance while maintaining throughput for other devices, thus preserving device state integrity without significantly impacting overall system productivity.
Solution Approach 2:
The patent maintains continuity of useful I/O operations by allowing traffic to continue flowing to and from devices that are not under maintenance. By selectively suspending traffic only to devices requiring maintenance, the system preserves overall productivity while ensuring reliability of device state for those being serviced.
Data Source
AI summary
A method comprising: sending an I/O message to an I/O device on an I/O bus, wherein the I/O device comprises an I/O device state, wherein the first I/O message is a I/O traffic type; receiving a LSR suspend request comprising a suspend instruction to suspend the I/O traffic type, an expected impact, and an I/O device identifier associated with the I/O device; setting the I/O device state to suspended in response to the receiving the LSR suspend request, wherein no I/O messages of the I/O traffic type are sent to the I/O device while the I/O device state is suspended; receiving a LSR resume request comprising a resume instruction to resume the I/O traffic type, the expected impact, and the I/O device identifier; setting the I/O device state to online in response to the receiving the LSR resume request; and sending a second I/O message to the I/O device.