Ingress Traffic Optimization via Site-Specific VM Grouping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Active/active data centers face challenges in ingress traffic optimization due to site-specific routing domains and the need for proprietary hardware in existing solutions like LISP, which complicates virtual machine migration and leads to suboptimal ingress paths.
Innovation Solution
A method for ingress traffic optimization in active/active data centers involves creating site-specific grouping constructs for virtual machines, dynamically updating network policies to advertise VM IP addresses, and using dynamic routing to ensure optimal path routing, independent of subnet and location.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If virtual machines are migrated between sites in active/active data centers, then virtual machine mobility and load balancing are improved, but ingress traffic paths become suboptimal due to site-specific routing domains
Solution Approach 1:
The patent segments the routing domain into site-specific groups, where each site maintains its own routing domain and adjacency to the local Internet peer. This segmentation allows VMs to be migrated between sites while each site independently optimizes its own ingress traffic paths, resolving the contradiction between VM mobility and traffic path optimization.
Solution Approach 2:
The patent applies local quality by allowing each site to have site-specific routing policies and advertisements. When a VM migrates, the source site withdraws its route advertisement and the destination site advertises the VM's IP address locally, ensuring that ingress traffic is optimally routed to the current site without requiring global routing changes.
2Ease of operation
If LISP ingress tunnel techniques are used to optimize ingress traffic, then traffic routing is improved, but device complexity increases due to required proprietary hardware line cards
Solution Approach 1:
The patent extracts the complex LISP hardware requirements and replaces them with standard routing protocol implementations. By using conventional routing protocol advertisements and withdrawals at border routers, the solution achieves ingress traffic optimization without requiring proprietary LISP line cards or specialized hardware, thus reducing device complexity while maintaining routing effectiveness.
3Reliability
If site-specific routing domains are implemented for each data center site, then local Internet peer adjacency is improved, but virtual machine migration between sites becomes complicated
Solution Approach 1:
The patent implements dynamic routing advertisements that automatically update when VMs migrate between sites. The system dynamically withdraws route advertisements from source sites and advertises them at destination sites using standard routing protocols, enabling seamless VM migration while maintaining site-specific routing domains and local Internet peer adjacencies without manual intervention.
Data Source
AI summary
A method of performing ingress traffic optimization for active/active data centers. The method creates site-specific grouping constructs for virtual machines that run applications that are advertised to the external networks. The site specific grouping constructs provide an abstraction to decouple virtual machines from traditional networks for common ingress network policies. Each site-specific container includes a list of the virtual machines currently located at the site as well as a unique identifier of the site. Each virtual machine in a container is identified through the abstraction of metadata tag, logical data center objects, or the virtual machine's unique name. The IP address of each virtual machine is retrieved from the guest operating system and a network policy is generated to advertise the IP addresses of the virtual machines to the site's routing peer.