Placement Service Module for Core Router Workload Balancing

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Modern high-end core routers face challenges in efficiently managing and distributing workload across multiple processing nodes with independently operating systems, leading to potential imbalances and inefficiencies in resource utilization, especially in large-scale service provider networks.

Innovation Solution

The implementation of a placement service module that dynamically manages and monitors processing nodes by making placement decisions based on predetermined policies, utilizing a placement engine to distribute processes across nodes based on resource availability and priority, ensuring balanced workload distribution across the system.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If multiple processing nodes are used to handle workload, then system capacity and scalability are improved, but workload imbalance and resource utilization inefficiency occur

Engineering Contradiction:

Improvesystem capacityVSAvoidworkload distribution

Core Design Contradiction:

ProductivityVSEase of operation

Solution Approach 1:

The patent implements a feedback mechanism where the placement service continuously monitors resource utilization metrics (CPU, memory, I/O) from each processing node and uses this information to dynamically adjust process placement decisions, ensuring balanced workload distribution while maintaining high system capacity

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system transitions from static process placement to dynamic placement where processes can be migrated between processing nodes based on real-time resource conditions. The placement service evaluates current node states and redistributes workloads dynamically to maintain optimal utilization across all nodes

Inventive Principle:

Principle #15Dynamics

2Adaptability or versatility

If processes are distributed across multiple nodes, then system scalability is improved, but complexity of managing process placement increases

Engineering Contradiction:

Improvesystem scalabilityVSAvoidprocess management complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent introduces a placement service as an intermediary component that manages process placement across multiple processing nodes. This centralized service simplifies complexity by providing a unified interface for process distribution, handling resource monitoring, and making placement decisions, thereby enabling scalability without proportionally increasing management complexity

Inventive Principle:

Principle #24Intermediary (Mediator)

Solution Approach 2:

The system architecture segments process management functions into distinct components: the placement service for decision-making, resource monitoring for data collection, and process migration for execution. This segmentation allows each component to handle specific aspects of scalability independently, reducing overall management complexity while maintaining system scalability

Inventive Principle:

Principle #1Segmentation

3Productivity

If resource utilization is optimized by consolidating workloads, then efficiency is improved, but system reliability and availability may deteriorate

Engineering Contradiction:

Improveresource utilization efficiencyVSAvoidsystem availability

Core Design Contradiction:

ProductivityVSReliability

Solution Approach 1:

The patent employs threshold-based parameter changes where the placement service monitors resource utilization metrics and triggers process migration when predefined thresholds are exceeded. This dynamic adjustment optimizes resource utilization by consolidating workloads during normal conditions while automatically redistributing them to maintain reliability when nodes become overloaded or fail

Inventive Principle:

Principle #35Parameter changes

4Ease of operation

If dynamic process placement is implemented, then workload balance is improved, but system complexity and overhead increase

Engineering Contradiction:

Improveworkload balanceVSAvoidsystem overhead

Core Design Contradiction:

Ease of operationVSDevice complexity

Solution Approach 1:

The placement service implements partial dynamic placement by monitoring all processing nodes but only migrating processes when necessary based on resource thresholds. This approach achieves adequate workload balance without the excessive overhead of continuous real-time migration, optimizing the trade-off between workload balance and system complexity

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS8850020B2Resource aware parallel process distribution on multi-node network devices

Publication Date: 2014.09.30 CISCO TECHNOLOGY INC

AI summary

Techniques are provided for monitoring a plurality of independently operating processing nodes in a core data packet routing system, wherein each processing node is running its own operating system instance. A placement decision is made for respective processes on selected ones of the independently operating processing nodes based on a predetermined policy, and then the respective processes are placed, based on the placement decision, on the selected ones of the independently operating processing nodes such that the respective processes become placed processes. A list of placed process/processing node pairs specifying a pairing of the placed processes and the processing nodes on which the placed processes have been placed is maintained so that a user can review the then-current set of process placements.