With the digitalization of the real economy, everything can be sensed, connected, and intelligent. The pure public cloud model cannot solve all the problems that occur during the digitalization of real economy due to constraints on real-time performance, bandwidth, security, and privacy. In this context, edge computing has been redefined, and related application scenarios have gradually emerged.
Ever since the concept of edge computing was introduced in 1997, IT engineers have paid more attention to network architecture and traffic optimization. After cloud providers entered this domain in 2016, edge computing was redefined as an extension of public cloud capabilities that support enterprise businesses with edge-cloud synergy architecture.
In the Cloud 2.0 era, enterprises in traditional verticals are exploring digital transformation. They are faced with many challenges, such as how to migrate services to the cloud, how to ensure service agility, and how to meet local service interaction and supervision requirements. An edge computing–oriented solution is one of the key enabling technologies needed to meet these transformation requirements.
Difficulty collecting device data
In vertical industries, diverse device interfaces and protocols cannot be unified in a short time. Data cannot be directly connected to the cloud due to constraints of service scenarios and industry supervision. Therefore, data collection, protocol adaptation and access, data preprocessing, and local service capabilities need to be implemented on enterprise service sites.
Difficulty integrating applications
Enterprises have constructed a large number of service systems to meet the increasing demands of applications. The complexity of enterprise services dictates the need for traditional applications, cloud applications, SaaS, multiple data sources, and multiple deployment areas. Traditional point-to-point integration results in complex application integration and dependency, as well as difficult architecture extension and maintenance.
Difficulty innovating with closed architectures
The existing traditional systems use closed architectures, and data is independent and hard to converge between these systems. As a result, it is difficult to realize architecture openness and convergence or use the latest container, data analysis, and AI technologies to support rapid service innovation.
Difficulty adapting to different scenarios with pure cloud architecture
In the process of digital transformation of vertical industries, enterprises seek to use an abundance of services on the public cloud. However, pure public cloud architecture often cannot solve problems such as real-time service performance, bandwidth optimization, data privacy protection, and security protection.
Edge-Cloud Video Analysis
Edge-Cloud Application Integration
Edge-Cloud IoT Awareness
Edge-Cloud Container Cluster
In the edge-cloud synergy architecture, edge-side video pre-processing through real-time surveillance can predict incidents, take appropriate actions, monitor situations, and track results.
Based on user scenarios, edge intelligence capabilities can be
customized and intelligent algorithms can be quickly deployed to reduce upload traffic and latency.
With integration capabilities deployed on- and off-cloud, messages, APIs, and data can be integrated through an on-and-off-cloud integration bus.
Interrupted data transmission can be resumed, greatly simplifying the architecture and dependency of enterprise application integration.
You can integrate data locally because data integration capabilities of the ROMA platform can be deployed on edge nodes.
IoT services can be preprocessed and closed at the edge on
premises to relieve the burden of the IoT platform. The edge cloud synergy architecture ensures service compliance and real-time performance.
The edge-cloud synergy architecture provides unified container cluster management, application deployment, and traffic distribution, eliminating single cluster failures.
This solution is ideal for multiple scenarios, such as service
traffic sharing, separation of services and data, separation of R&D and production, and decoupling of computing and services.
Architecture of the Edge-Cloud Synergy Solution
The edge-cloud synergy solution provides an open, cloud-native edge-cloud synergy architecture and extends extensive cloud services to edge nodes, enabling a synergy of containers, devices, application integration, and video service capabilities and allowing you to develop edge applications effectively.
In the solution, services are mainly processed at the edge. Devices, applications, and video data can be connected to the edge. Service capabilities such as IoT awareness, AI inference, application integration, and near-field computing can also be conducted at the edge based on service demand. The solution can be easily mapped to reference architectures of different vertical industries to preprocess data or close services locally.
The edge nodes and services involved are centrally managed in the cloud, enabling registration, discovery, configuration management, service delivery, and O&M information reporting of edge nodes.
The solution supports enterprise edges and edge sites such as CDN and distributed data centers at the edge. It supports HUAWEI CLOUD and the HCS Online solution, which is an on-premises extension of HUAWEI CLOUD in the cloud.
Decoupled and Open Architecture
The cloud-native edge-cloud synergy architecture supports registration of devices embedded with heterogeneous hardware such as x86 servers, ARM, NPUs, and GPUs as edge nodes. You can easily deploy, update, and iterate services on edge nodes by packing services into containers.
Deppon ExpressHUAWEI CLOUD helped Deppon deploy AI models on sites based on the edge-cloud synergy architecture, enabling automatic extraction of service-related information through optical character recognition (OCR) and effective identification of violent sorting behavior.
Jingying ShuzhiHUAWEI CLOUD helped Jingying Shuzhi build an intelligent video identification system for coal mines based on device-edge-cloud synergy. The system monitors and manages prospecting drainage, automatically identifies the number of unloading poles, and calculates drilling depth to prevent falsification of engineering materials and eliminate drainage prospection faults.