Microservices Architecture for Dynamic Network Function Allocation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data center architectures face challenges in efficiently managing and optimizing network functions, leading to potential performance bottlenecks due to hardcoded network functions and limited resource utilization.
Innovation Solution
The implementation of a microservices architecture that allows for the dynamic discovery and allocation of optimized resources, such as FPGAs and hardware accelerators, to perform specific network functions, thereby enhancing flexibility and speed.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If hardcoded network functions are used in traditional data center architecture, then device simplicity is maintained, but network function flexibility and resource utilization efficiency deteriorate
Solution Approach 1:
The patent segments network functions from hardware devices by introducing virtualization technology. Virtual network functions (VNFs) are extracted from physical network devices and can be independently deployed, managed, and scaled on standard hardware platforms. This segmentation enables flexible network function deployment while maintaining relatively simple underlying hardware infrastructure.
Solution Approach 2:
The patent implements universal hardware platforms that can host multiple different network functions through virtualization. A single physical server can run multiple virtual machine instances, each providing different network functions (firewall, load balancing, routing, etc.). This multi-functionality approach replaces the traditional one-device-one-function model, improving resource utilization while maintaining device simplicity.
2Productivity
If traditional monolithic network devices are used, then device complexity is reduced, but resource utilization efficiency and performance optimization deteriorate
Solution Approach 1:
The patent introduces dynamic resource allocation mechanisms where network functions can be dynamically instantiated, migrated, and scaled based on actual traffic demands. Virtual machine instances can be dynamically created or terminated, and resources can be dynamically reassigned between different network functions, enabling high resource utilization efficiency while managing complexity through automation.
Solution Approach 2:
The patent introduces a virtualization layer as an intermediary between physical hardware and network functions. This virtualization layer abstracts and manages the complexity of hardware resources, presenting simplified interfaces for deploying and managing network functions. The intermediary handles resource allocation, scheduling, and optimization, improving productivity while containing device architecture complexity within the virtualization layer.
3Adaptability or versatility
If network functions are tightly coupled to specific hardware, then ease of operation is improved, but adaptability and scalability deteriorate
Solution Approach 1:
The patent enables network functions to be copied and replicated across multiple virtual machine instances. A single network function can be instantiated multiple times on different physical or virtual hosts, enabling horizontal scaling. The virtualization layer manages these copies transparently, maintaining ease of operation through standardized deployment interfaces while achieving high scalability.
Solution Approach 2:
The patent adds a virtualization dimension to the traditional hardware-network function mapping. Instead of direct one-to-one mapping, network functions are deployed in a virtual dimension that can be independently scaled and managed. This additional abstraction layer enables scaling across multiple dimensions (physical hosts, virtual clusters, geographic locations) while maintaining operational simplicity through unified management interfaces.
Data Source
AI summary
A computing apparatus, including: a hardware computing platform; and logic to operate on the hardware computing platform, configured to: receive a microservice instance registration for a microservice accelerator, wherein the registration includes a microservice that the microservice accelerator is configured to provide, and a microservice connection capability indicating an ability of the microservice instance to communicate directly with other instances of the same or a different microservice; and log the registration in a microservice registration database.


