Parallel Serverless Deployment for Telecommunication Applications
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Deploying telecommunication applications in server-based systems using microservices leads to inefficient resource utilization, increased costs, and decreased responsiveness due to the need for additional servers and infrastructure, especially with the rise in network functions and services, making it challenging to scale and manage resources effectively.
Innovation Solution
A cloud-agnostic serverless framework that utilizes parallel deployment of telecommunication applications across clusters of computers, leveraging Functions as a Service (FaaS) and lightweight threads to dynamically allocate resources, eliminating the need for dedicated servers and enabling flexible scaling, thereby improving resource utilization and reducing infrastructure costs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If server-based systems with microservices are used to deploy telecommunication applications, then the applications can be managed and deployed, but resource utilization becomes inefficient and infrastructure costs increase
Solution Approach 1:
The patent uses containerization to create lightweight, portable copies of application environments that can be deployed across multiple hosts without requiring full server installations. Containers package applications with their dependencies, enabling efficient resource utilization while maintaining deployment capability
Solution Approach 2:
The orchestration system provides universal management capabilities across heterogeneous cloud infrastructures. A single orchestration platform can deploy and manage applications on different cloud providers and infrastructure types, eliminating the need for separate server management systems and improving resource utilization
2Power
If additional dedicated servers are added to accommodate increased network functions and services, then processing capacity and storage increase, but infrastructure costs and system complexity increase
Solution Approach 1:
The patent combines multiple application instances and their required resources into shared infrastructure environments. Containers from different applications can share the same underlying server resources, allowing processing capacity to scale without proportionally increasing infrastructure complexity
Solution Approach 2:
The orchestration system dynamically allocates and scales computing resources based on actual application demands. Resources can be dynamically assigned to different applications as needed, providing scalable processing capacity without permanent infrastructure commitments
3Reliability
If servers are dedicated to specific cloud types or customized for specific clouds, then cloud-specific requirements are met, but cloud agnosticism and flexibility are reduced
Solution Approach 1:
The orchestration system serves as a universal layer that abstracts cloud-specific details from application deployment. It can manage applications across multiple cloud providers (AWS, Azure, Google Cloud, private clouds) using standardized container interfaces, maintaining both cloud-specific compatibility and cloud agnosticism
Data Source
AI summary
A method for parallel deployment of telecommunication applications may be provided. The method, executed by a cloud adaptor, may include receiving a network topology, a resource status of each of one or more available clusters, and a cluster context associated with the each of the one or more available clusters. The method may include spawning a plurality of threads wherein each of the plurality of threads may be associated with a respective cluster and a respective task to be executed on the respective cluster, and wherein the respective task may be associated with a respective function to be performed. The method may include setting the cluster context associated with the respective cluster to the respective function to be performed associated with the respective task to be executed on the respective cluster; and directing parallel execution of the plurality of threads across the one or more available clusters.