Adaptive Credit-Based Flow Control for Information Handling Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Credit-based flow control systems in information handling systems face limitations such as overhead in the request/grant approach, inability to reclaim credits, and lack of buffer sharing across ports, leading to inefficiencies and potential congestion.
Innovation Solution
An adaptive credit-based flow control system that dynamically allocates shared buffer pools across ports, allowing for real-time monitoring of sender queue sizes and intelligent credit allocation based on demand, enabling buffer sharing and reducing latency by eliminating the need for a request/grant mechanism.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If credit-based flow control uses a request/grant approach with fixed buffers per port, then flow control can be implemented, but overhead increases and buffer sharing across ports becomes impossible
Solution Approach 1:
The patent merges previously separate per-port buffer allocations into a shared buffer pool that serves multiple ports. The flow control entity dynamically allocates buffers from this shared pool based on real-time需求的, eliminating the need for separate fixed buffers at each port while reducing overhead through centralized management.
Solution Approach 2:
The system transitions from static fixed buffer allocations to dynamic buffer allocation. The flow control entity continuously monitors buffer usage and demand across ports, reallocating buffers dynamically to match actual needs, thereby reducing overhead while maintaining reliable flow control.
2Productivity
If credits are sent to sender indicating available buffer space, then sender can transmit data, but credits cannot be taken back once allocated
Solution Approach 1:
The system implements feedback mechanisms where the flow control entity continuously monitors actual buffer usage and demand. When buffers are not fully utilized or demand decreases, the system can reclaim credits and reallocate them, creating a closed-loop control system that adapts to changing conditions rather than maintaining static credit allocations.
Solution Approach 2:
The credit allocation system becomes dynamic rather than static. Credits can be allocated, monitored, and reclaimed based on real-time buffer usage patterns and demand fluctuations, allowing the system to optimize data transmission while preventing credit wastage through automated reclamation when buffers become available.
3Reliability
If fixed buffer pools are allocated to each port, then flow control can be implemented, but buffer sharing across ports is not possible
Solution Approach 1:
The flow control entity serves multiple ports universally from a shared buffer pool rather than dedicating separate buffers to each port. This universal approach allows any port to access any buffer in the pool based on demand, enabling buffer sharing across multiple ports while maintaining reliable flow control through centralized management.
Solution Approach 2:
Previously separate per-port buffer pools are merged into a single shared buffer pool that serves all ports. This consolidation enables buffer sharing across ports while the flow control entity manages allocations dynamically, maintaining flow control reliability through unified oversight rather than isolated per-port management.
4Reliability
If credits are allocated based on fixed buffer availability, then flow control can be implemented, but additional credits cannot be allocated when shared buffer pool has congestion
Solution Approach 1:
The system transitions from static credit allocation based on fixed buffer pools to dynamic credit allocation based on real-time monitoring of shared buffer pool status and demand. The flow control entity can allocate additional credits when buffers are available and reclaim them when congestion occurs, providing flexible credit allocation that adapts to changing network conditions.
Solution Approach 2:
The flow control entity implements continuous feedback monitoring of buffer usage and demand across the shared pool. Based on this feedback, the system dynamically adjusts credit allocations, allowing additional credits when resources are available and reclaiming credits when congestion occurs, thereby balancing flow control reliability with allocation flexibility.
Data Source
AI summary
The present invention relates generally to an information handling system. Aspects of the present invention include an adaptive credit-based flow control in an information handling system. In embodiments, a request/grant credit system can be eliminated since the receiver can dynamically allocate credits based on demand at the sender. In embodiments, the sender can provide information related to its queue size to the receiver. The receiver can estimate queue size and demand based on the estimated queue size. In embodiments, the receiver allocates credits based on sender demand.