Dynamic Allocation Of Service Components Of Information Service In Hierarchical Telecommunication Architecture

Link to the online patent PDF

Overview

Embodiments relate to allocating resources of computing devices for providing information service in a network. The computing devices may be hierarchically structured and may include, for instance, cloud servers, telecommunication servers, edge gateways, and client devices. A system environment may include a hierarchical orchestrator coordinating with one or more local orchestrators to allocate service components (for example, a discrete functional software or hardware component) to computing devices. The orchestrators can automatically reallocate resources responsive to detecting update events such as a change in traffic or payload on the network.

The Problem

Traditional cloud computing architectures, while powerful, face significant limitations in use cases that demand low latency and high reliability. When a centralized data center experiences high traffic or workload, its response time (latency) increases, leading to poor performance. For modern applications like autonomous vehicles, augmented reality, and real-time monitoring, this lag is unacceptable. The core challenge is to effectively distribute application components (services) across a wide range of computing devices—from massive cloud servers to small edge devices—while maintaining high performance and reliability.

The Solution

The patent proposes a hierarchical system that dynamically allocates service components across different levels of a telecommunication network. This architecture consists of multiple tiers of computing devices, including cloud servers, telecommunication servers, edge devices, and gateways.

Two key components manage this allocation:

1. Hierarchical Orchestrator: A central controller that has a global view of the entire network. It makes high-level decisions about where to place service components based on a set of rules, considering factors like cost, performance requirements, and resource availability.
2. Local Orchestrators: These operate at specific levels of the hierarchy (e.g., managing a group of edge devices) and handle the real-time allocation and execution of service components within their local domain.

This system can automatically detect changes in the network (like a spike in traffic or a device failure) and dynamically re-allocate service components to optimize performance in real-time.

Why It Matters

This technology is a critical enabler for the next generation of digital services. By moving computation closer to the end-user (to the "edge"), it dramatically reduces latency and improves reliability. This is essential for:

• Real-time Applications: Services like connected cars, drone control, and factory automation require near-instantaneous response times that traditional cloud models cannot guarantee.
• Network Efficiency: It reduces the amount of data that needs to be sent back to a central cloud, easing network congestion and lowering operational costs.
• Scalability and Resilience: The system can automatically adapt to changing demands and recover from failures by redistributing workloads, creating a more robust and scalable infrastructure.

Ultimately, this patent provides a blueprint for building intelligent, self-adapting networks capable of supporting the demanding requirements of the future digital world.

Relevance Beyond Telecommunications

While the patent is framed within a telecommunications context, its core principles of hierarchical orchestration and dynamic resource allocation are broadly applicable to any complex, distributed system that requires a blend of centralized control and localized autonomy. Here are a few examples:

• Smart Cities: City-wide infrastructure, including traffic management systems, public safety cameras, and smart utility grids, can be managed using this model. Local orchestrators could manage resources within a specific district or neighborhood, while a central, hierarchical orchestrator optimizes city-wide operations based on real-time data (e.g., rerouting traffic during a major event).
• Industrial IoT (IIoT) and Manufacturing: In a smart factory, this architecture can orchestrate the complex web of machinery, robotic arms, and quality control sensors. A local orchestrator might manage a single production line, ensuring low-latency control, while a factory-wide orchestrator optimizes the overall production schedule, supply chain inputs, and energy consumption.
• Healthcare: The system can be applied to real-time patient monitoring. Wearable sensors and bedside devices (the edge) can process vital signs locally to provide immediate alerts. This data can then be aggregated and sent to hospital-level servers (local orchestration) or a regional cloud (hierarchical orchestration) for long-term analysis and record-keeping, ensuring both immediate response and large-scale data insights.
• Retail and Logistics: Large retail chains can manage in-store technology (like smart shelves, POS systems, and foot-traffic cameras) with local orchestrators in each store. A central orchestrator at the corporate level could then analyze data from all stores to manage inventory, optimize supply chains, and make strategic business decisions.

The fundamental concept of placing computational resources and decision-making at the most effective point in a hierarchy is a powerful paradigm for building efficient, scalable, and resilient systems in virtually any industry.

Technical Details

The system is composed of several key elements organized in a flexible hierarchy:

• Computing Hierarchy: The architecture is multi-layered, typically including:
  • Level 1: Cloud Servers: Centralized data centers for heavy computation and large-scale data storage.
  • Level 2: Telecommunication Servers: Intermediate servers within the telecom provider's network.
  • Level 3: Edge Devices: Devices like edge routers, base stations, or local micro-data centers located close to the user.
  • Level 4 & 5: Gateways and Client Devices: User-premise equipment and end-user devices.
• Orchestration:
  • The Hierarchical Orchestrator uses a set of rules and parameters to make strategic allocation decisions. These rules can be based on latency requirements, cost, security policies, or geographic location.
  • Local Orchestrators monitor resources within their domain (e.g., a cluster of edge servers) and execute the deployment instructions sent by the hierarchical orchestrator. They also report status and alert the central orchestrator to local events.
• Dynamic Allocation Process:
  1. Detection: The system continuously monitors the network for "update events," such as changes in traffic, device availability, or new service requests.
  2. Re-evaluation: Upon detecting an event, the Hierarchical Orchestrator automatically updates its allocation plan based on the current state of the network and its set of rules.
  3. Instruction: New instructions are sent to the relevant Local Orchestrators.
  4. Execution: The Local Orchestrators then deploy, decommission, or move service components to the newly assigned computing devices.

Status: Issued

Application Number: 15/922,817

Patent Number: 10693704

Filing Date: 2018-03-15

Issue Date: 2020-06-23

‍