Topology-Centric Resource Management for Service Clusters

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Large-scale service clusters face challenges in optimizing network topology and resource management due to physical limitations of switches and varying bandwidth, leading to suboptimal availability and performance.

Innovation Solution

Implementing topology-centric resource management by creating an optimized network topology based on application flows and bandwidth requirements, using a service graph to assign switches and optimize load balancing, and employing centralized or decentralized monitoring schemes to maintain the topology view.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Quantity of substance

If a hierarchical topology with multiple network switches is used to connect service modules, then the cluster can accommodate more nodes and expand capacity, but the network complexity increases and bandwidth is limited by switch uplinks

Engineering Contradiction:

Improvenumber of nodesVSAvoidnetwork topology complexity

Core Design Contradiction:

Quantity of substanceVSDevice complexity

Solution Approach 1:

The patent segments the service cluster into multiple service modules, each with its own dedicated switch. This segmentation allows the system to scale by adding modules independently while maintaining manageable complexity through modular architecture. Each service module is a self-contained unit with dedicated resources.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a dedicated switch as an intermediary component between service modules and the core network. This dedicated switch acts as a mediator that handles all traffic for its associated service module, isolating it from other modules and simplifying the overall network topology management.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Quantity of substance

If a hierarchical topology with multiple network switches is used to connect service modules, then the cluster can accommodate more nodes and expand capacity, but the bandwidth is limited by the uplink of each switch

Engineering Contradiction:

Improvenumber of nodesVSAvoidbandwidth

Core Design Contradiction:

Quantity of substanceVSPower

Solution Approach 1:

The patent segments the network bandwidth allocation by assigning a dedicated switch to each service module, creating dedicated bandwidth paths. This segmentation ensures that bandwidth is allocated specifically to each module's needs rather than sharing limited uplink capacity, allowing scalable bandwidth provision as nodes are added.

Inventive Principle:

Principle #1Segmentation

3Adaptability or versatility

If traffic is routed through multiple switches in a hierarchical topology, then nodes can be connected across non-adjacent cages, but the availability decreases due to potential switch failures

Engineering Contradiction:

Improvenetwork connectivityVSAvoidavailability

Core Design Contradiction:

Adaptability or versatilityVSReliability

Solution Approach 1:

The patent segments the network into isolated service module domains, each with its own dedicated switch. This segmentation means that a failure in one service module's switch does not propagate to other modules, as each module is isolated in its own switching domain. This modular isolation significantly improves system availability.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The dedicated switch acts as an intermediary that isolates traffic for its associated service module from other modules. This intermediary component provides a containment boundary that prevents failure propagation, ensuring that issues in one module do not affect the availability of other modules.

Inventive Principle:

Principle #24Intermediary (Mediator)

4Reliability

If a dedicated switch is assigned to each service module, then availability is improved by isolating failures, but the number of switches and device complexity increases

Engineering Contradiction:

ImproveavailabilityVSAvoidnumber of switches

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent applies multi-functionality by having the dedicated switch serve multiple purposes: it acts as a layer 2 switch for service module connectivity, a layer 3 router for inter-module communication, and a bandwidth manager for dedicated capacity allocation. This multi-functionality reduces the need for additional specialized devices while maintaining high availability.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS7894372B2Topology-centric resource management for large scale service clusters

Publication Date: 2011.02.22 IAC SEARCH & MEDIA INC

AI summary

Topology-centric resource management for large scale service clusters is described herein. According to certain embodiments of the invention, techniques include 1) creating optimized topology with network switches to connect service modules based on application flows and bandwidth requirements, 2) providing centralized or decentralized monitoring schemes to maintain the topology view of a service cluster, and 3) using the topology information for optimizing load balancing and service information dissemination. Other methods and apparatuses are also described.