Virtual Output Queue Rate Control via Egress Port Feedback

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In computer networks, virtual output queues (VoQs) lack awareness of congestion states in egress port queues they are linked to, leading to potential overload and increased latency, resulting in dropped packets due to inaccurate regulation methods based solely on egress port queue depth.

Innovation Solution

Implement a dynamic control loop that adjusts VoQ output rates based on egress port queue transmit rates, depth, and available shared resources, using a destination node's monitor engine, usage engine, and rate engine to determine utilization and backlog percentages, thereby modifying latency thresholds and output rates to adapt to network changes.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If VoQ sends traffic into the fabric based on estimated congestion state, then egress buffers can be kept occupied for appropriate QoS scheduling, but the egress port buffers and fabric node may be overloaded resulting in dropped data packets

Engineering Contradiction:
ImproveQoS scheduling efficiencyVSAvoidpacket drop rate
Core Design Contradiction:
ProductivityVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where the egress port queue depth is continuously monitored and used to dynamically adjust the VoQ output rate. The source node receives congestion information from the destination node and adjusts its transmission rate accordingly, creating a closed-loop control system that prevents buffer overload while maintaining QoS scheduling efficiency

Inventive Principle:
Principle #23Feedback

Solution Approach 2:

The patent makes the VoQ output rate dynamic by adjusting it based on real-time egress port queue depth information. Instead of using a fixed or purely estimated rate, the system continuously adapts the transmission rate to match the actual congestion state of the egress queue, preventing both overload and underutilization

Inventive Principle:
Principle #15Dynamics

2Device complexity

If regulation is based solely on egress port queue depth, then the control mechanism is simple, but it does not provide accurate depiction of egress port queue state leading to inappropriate output rate control

Engineering Contradiction:
Improvecontrol mechanism complexityVSAvoidcongestion state accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent introduces an intermediary feedback message that carries congestion information from the destination node to the source node. This intermediary mechanism provides accurate congestion state information without requiring complex direct communication or estimation algorithms at the source node, maintaining simplicity while improving measurement precision

Inventive Principle:
Principle #24Intermediary (Mediator)

3Stability of the object's composition

If VoQ output rate is not dynamically adjusted, then the system is stable and easy to implement, but it cannot adapt to sudden changes in available space on shared resources and oscillating service patterns

Engineering Contradiction:
Improvesystem stabilityVSAvoidnetwork adaptability
Core Design Contradiction:
Stability of the object's compositionVSAdaptability or versatility

Solution Approach 1:

The patent uses feedback from egress port queue depth to dynamically adjust the VoQ output rate, creating a stable yet adaptive control system. The feedback loop maintains system stability through controlled adjustments while enabling adaptability to changing network conditions, resource availability, and service patterns

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS10536385B2Output rates for virtual output queses
Publication Date: 2020.01.14 HEWLETT PACKARD ENTERPRISE DEV LP
  • US10536385B2 patent drawing
  • US10536385B2 patent drawing
  • US10536385B2 patent drawing

AI summary

Examples include sampling a transmit rate of an egress port queue on a destination node, determining a utilization percentage of the egress port queue based on the transmit rate and a total rate capacity of the egress port, and determining a backlog percentage of the egress port queue. Examples also include determining an output rate for a virtual output queue on a source node based on the utilization percentage and the backlog percentage. The virtual output queue is for the egress port queue.