Inter-Network Service Chaining for BMS and Virtual Workloads
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing network technologies struggle to efficiently steer traffic between bare metal servers and virtual execution elements through service nodes using Ethernet Virtual Private Networks (EVPN) as an underlying transport technology, while maintaining scalability and addressing overlapping address spaces.
Innovation Solution
A centralized controller constructs inter-network service chains by re-originating IP and EVPN routes to steer traffic through service nodes, using techniques such as MPLS over GRE/UDP and VXLAN, ensuring traffic is directed through service nodes before reaching its final destination, and extending the service chain to bare metal servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If traditional routing protocols are used to steer traffic between bare metal servers and virtual execution elements, then network scalability is maintained, but traffic cannot be efficiently directed through service nodes
Solution Approach 1:
The patent introduces a centralized controller as an intermediary between traditional routing protocols and service chain management. This controller translates high-level service chain policies into low-level routing instructions, enabling efficient traffic steering through service nodes without requiring complex modifications to existing routing infrastructure. The controller acts as a mediator that reconciles the simplicity of policy-based management with the complexity of underlying routing operations.
Solution Approach 2:
The patent segments the routing functionality into two distinct layers: a control plane handled by the centralized controller that manages service chain policies, and a data plane that executes traffic forwarding through service nodes. This segmentation allows traditional routing protocols to continue operating at the network layer while the controller independently manages service chain steering, thereby maintaining scalability while enabling efficient service node integration.
2Adaptability or versatility
If service chains are extended to include bare metal servers, then network flexibility is improved, but address space overlap issues arise
Solution Approach 1:
The centralized controller serves as an intermediary that manages address space allocation and translation for service chains involving bare metal servers. It maintains a global view of address spaces across virtual and physical networks, performing address translation and route manipulation to prevent overlaps. This intermediary function enables flexible service chain extensions while maintaining reliable address space management through centralized coordination.
Solution Approach 2:
The patent dynamically changes routing parameters such as next-hop addresses and route metrics based on the specific requirements of each service chain. By adjusting these parameters in real-time, the system can accommodate bare metal servers with overlapping address spaces, directing traffic through appropriate service nodes while maintaining address space integrity. This parameter manipulation allows flexible adaptation without compromising address space reliability.
3Productivity
If centralized control is implemented for service chain management, then traffic steering efficiency is improved, but control plane complexity increases
Solution Approach 1:
The centralized controller is designed with multi-functionality, handling service chain management, route manipulation, address space coordination, and policy enforcement through a single unified platform. By consolidating these diverse functions into one controller, the system achieves efficient centralized traffic steering without proportionally increasing complexity, as the controller leverages common infrastructure and algorithms across different management tasks.
Solution Approach 2:
The controller uses copying mechanisms to distribute service chain policies and routing information to relevant network elements without requiring complex real-time synchronization. By creating and propagating copies of routing tables and service chain configurations to edge devices and service nodes, the system achieves efficient traffic steering while keeping the control plane manageable through simple replication rather than complex coordination protocols.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Techniques are described in which a centralized controller constructs a service chain between a bare metal server (BMS) and a virtual execution element (e.g., virtual machine or container), or in some instances a remote BMS, across a plurality of networks. In some examples, the controller may construct a service chain between a BMS and a virtual execution element or remote BMS using Ethernet Virtual Private Network (EVPN) - Virtual Extensible Local Area Network (VXLAN) and Internet Protocol Virtual Private Networks (IP VPNs) such as BGP/Multiprotocol Label Switching (BGP/MPLS) IP VPNs.