TietoEVRY Scalable Edge Reference Platform
TietoEVRY edge reference platform, the open-source based solution demonstrated in this paper, is a viable alternative to existing commercial edge solutions to improve total cost of ownership (TCO) without compromising on performance. The platform, due to its flexibility, scalability, cloud-native characteristic and Intel technology is powerful for emerging edge use cases across various industries such as Telecom, Industry, Enterprise, IoT, SmartCities, SmartHomes, Automotive, or MedTech.
This paper is divided into 3 major parts.
- First (chapters I, II) discussing the problem, MEC (multi-access edge computing) evolution and the use case itself
- Second (chapters III and IV) explaining technical solution, deployment, benchmarking, and scalability
- Third (chapters V to VII) describing the advantages solution brings to the industry and business
MEC (Multi-access edge computing) and what it brings to the industry
Today’s booming demand for low latency edge applications creates the need for highly flexible, scalable, and automated solutions. Edge computing is transforming the way data is being handled, processed, and delivered from millions of devices globally. The explosive growth of internet-connected devices, fueled by fast networking (such as 5G), accelerate the creation of new use cases, such as video analytics, self-driving cars, artificial intelligence, robotics, and others.
Multi-workload scalable platform
The goal of this paper is to exemplify the application of TietoEVRY (which the company provides to the industry), based on extensive technology experience and strong partnerships, in building cutting edge solutions powered by Intel technology.
We selected the Open-source Intel Smart Edge Open (formerly known and referred to in this paper as OpenNESS) platform to create a reference implementation of a scalable edge computing platform. The example deployment comprises of real-time, AI inference video analytics solution for Smart Cities implementing an Automated Pedestrian Alert System (APAS).
OpenNESS exposes Intel hardware features to the Kubernetes based, containerized Edge environment and enables easy deployment and optimized orchestration (using OpenVINO™, Data Plane Development Kit (DPDK), Real-time kernel etc.) of various edge use cases such as media analytics, Content Delivery Network (CDN) up to 5G access, and core network functions.
Smart City - pedestrian safety use case
Urbanization increases the number of people and vehicles on the move in metropolitan areas. As road traffic increases, so does the risk of traffic accidents, especially at intersections. The risk of injury and death is especially high for pedestrians. APAS is an IoT solution to improve pedestrian safety when crossing (or intending to cross) the street by alerting incoming cars about the potential risk of a person on the crossroad via the car’s cockpit warnings. This is especially crucial for multi-lane crossroads (where stopped cars may limit another driver’s zebra crossing visibility) or in bad weather conditions.
The successful implementation of open and intelligent edge computing solutions relies on the deployment of commercial-off-the-shelf (COTS), white-box hardware closer to sensors, devices, and end-users. These edge computing platforms need to deliver high-performance processing while providing sustainable operations in outdoor or semi-outdoor environments, keeping a constant eye on power consumption. Advantech 5G Edge Servers balance the latest generation Intel® Xeon® Scalable processors with advanced AI and video acceleration in optimized, high density, high-reliability platforms designed to withstand edge environmental conditions.