As global 5G deployment accelerates, multi-access edge computing (MEC) is playing an essential role in helping to deliver on 5G commitments by significantly reducing latency, increasing connection speed, enhancing network security, and improving quality of service to end users. Leveraging the performance of 5G, GPU-based MEC solutions can effectively harness the power of Internet of Things (IoT), artificial intelligence (AI) and machine learning technologies to open up enormous opportunities with a wide array of applications including autonomous vehicles, connected cars, virtual reality (VR), augmented reality (AR), smart cities, smart transportation and smart manufacturing.
GPU-based 5G Edge Computing Solutions
In the era of 5G, MEC is an evolution of cloud computing, moving data processing from centralized data centers out to the network edge and closer to application end users. With the integration of MEC to effectively distribute networking demands, 5G’s tremendously improved capabilities can handle an exponentially increasing number of connected devices, and enable a wide range of latency- and bandwidth dependent, cross-industry technologies and applications. GPU-based MEC solutions further bring superior data and graphic processing to enable critical, high-load applications that have a need for accelerated computing and faster time-to-decision.
By leveraging more than 20 years of expertise in developing highly reliable and available embedded computing systems, ADLINK is a premier supplier of extensive, cost-effective COTS, as well as fast time-to-market ODM solutions to worldwide tier-one telecommunications equipment manufacturers (TEMs). ADLINK’s 1U/2U MEC servers MECS-6110 and MECS-7210, two new additions to its communication and networking product portfolio, are among the first platforms to fully comply with Open Telecom IT Infrastructure (OTII) defined by the Open Data Center Committee (ODCC). The two servers are both designed with dual full-height full-length (FHFL) PCIe expansion slots reserved for access to accelerated computing hardware such as GPU and FPGA. In one use case, a customer in the manufacturing industry deployed the MECS-7210 into the low-latency LTE/5G private network of its smart factory to effectively identify defective products and improve operational efficiency through a GPU-based, AI-enabled application.