Virtual Route Controller Bypasses Gateway for VPC Latency

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current cloud service provider infrastructure faces challenges in handling increased bandwidth and port density requirements for virtual private clouds (VPCs), leading to high latency, decreased per connection consistency, and high management costs due to the need for extensive routing tables and expensive routers, especially with the use of x86 systems for network function virtualization.

Innovation Solution

Implementing an offload controller (OC) and a programmable switch ASIC (CSW) to bypass the gateway, allowing for direct packet forwarding between customer premises equipment (CPE) and virtual machines (VMs) within VPCs, using a VPC controller and virtual route controller (VRC) to manage routing and configure routes, thereby reducing latency and energy usage while increasing connection consistency without relying on virtual routing and forwarding (VRF).

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If all packets are handled at the gateway, then routing control is centralized, but network latency increases and per connection consistency decreases

Engineering Contradiction:

Improveper connection consistencyVSAvoidnetwork latency

Core Design Contradiction:

ReliabilityVSLoss of time

Solution Approach 1:

The patent segments the network architecture by introducing customer switches at the customer premises that handle local packet forwarding independently from the gateway. This divides the previously centralized gateway function into distributed components (customer switches and gateway), allowing packets to be forwarded locally without always traversing the gateway, thus reducing latency while maintaining routing control through coordinated route management.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces customer switches as intermediary devices between the gateway and customer premises equipment. These switches act as mediators that can locally manage packet forwarding decisions, reducing the need for packets to traverse the entire path to the gateway and back, thereby decreasing network latency while maintaining connection consistency through coordinated route management.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If x86 systems are used for network function virtualization, then flexibility and ease of deployment improve, but power consumption increases and performance is limited

Engineering Contradiction:

Improveease of deploymentVSAvoidpower consumption

Core Design Contradiction:

Ease of operationVSUse of energy by stationary object

Solution Approach 1:

The patent changes the hardware parameter from general-purpose x86 systems to specialized network switch hardware. This parameter change optimizes power efficiency and performance for network switching functions while maintaining ease of deployment through standardized hardware interfaces and automated configuration management by the gateway.

Inventive Principle:

Principle #35Parameter changes

3Reliability

If VRF is deployed for each VM host, then tenant traffic isolation is achieved, but the number of routing tables proliferates and management costs increase

Engineering Contradiction:

Improvetenant traffic isolationVSAvoidnumber of routing tables

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent merges routing table management by allowing customer switches to share routing information and use common routing tables for multiple tenants. Instead of each VM host maintaining separate VRF routing tables, the customer switches collectively manage routing information, reducing the total number of routing tables while maintaining tenant isolation through switch-level enforcement and coordinated route management.

Inventive Principle:

Principle #5Merging (Combining)

4Adaptability or versatility

If gateway handles all overlay network traffic, then centralized control is maintained, but port density and bandwidth requirements cannot be met

Engineering Contradiction:

Improveport densityVSAvoidgateway capacity requirements

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent segments the traffic handling function by introducing customer switches that handle local packet forwarding independently. This segmentation distributes the bandwidth and port density requirements from the centralized gateway to multiple distributed customer switches, allowing the system to meet high port density and bandwidth requirements without overloading the gateway.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent adds a new dimension to the network architecture by introducing a customer switch layer between the gateway and customer premises equipment. This dimensional addition allows traffic to be handled at multiple levels (gateway for routing decisions, customer switches for local forwarding), enabling the system to scale to meet high port density and bandwidth requirements while maintaining centralized routing control.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

Data Source

PatentUS10855584B2Client-equipment-peering virtual route controller

Publication Date: 2020.12.01 CLOUD INTELLIGENCE ASSETS HOLDING (SINGAPORE) PTE LTD

AI summary

The techniques discussed herein include an offload controller, a virtual routing controller, and/or virtual routing objects. In some instances the virtual routing controller may be configured and/or positioned to peer client premises equipment (CPE). This may include establishing an external gateway protocol session with CPE and generating a virtual routing object based at least in part on the session. In some examples, this virtual routing object may be used to configure routes between a switch and a virtual private cloud (VPC) and/or virtual machine (VM).