Eagle-Lanner tech blog

 

Sven Freudenfeld, Chief Technology Officer for the Telecom Application BU at Lanner Electronics, joins Intel® Chip Chat Network Insights to discuss the evolution of SD-WAN and uCPE platform, and share how Lanner is focused on defining new platform architectures that scale in the SD-WAN, uCPE and MEC use cases.

The rapid growth of cloud data centers in recent years has been the driving force for the introduction of single-lane 25 Gigabit Ethernet specification, and this new technology has been the game changer for switch and server designs. The 25G Ethernet, firstly announced by 25G Ethernet Consortium with members of Arista, Broadcom, Google, Mellanox Technologies and Microsoft in 2014, has since revolutionized Ethernet development roadmap.

When 4G/LTE was first introduced, communication service providers (CSPs) were delivering uniform services and relying on proprietary equipments when they expanded their coverage. However, CSPs have been challenged by the widespread of mobile devices and cloud applications, which has led to the era of Big Data. CSPs revenue has shrunk due to a highly competitive market while users demand better quality of service (QoS) and quality of experience. In order to improve competitiveness in the demanding market, CSPs have embraced NFV (network function virtualization) to capture future business opportunities.

There is no doubt that SD-WAN has become one of the most widespread paradigm shift in the technology field over the past few years. Indeed, SD-WAN offers multiple advantages for enterprise WAN management, including the simplicity in traffic routing, lowered dependence on MPLS, and the cost-effectiveness through implementation on existing traditional WAN hardware. However, like all the widespread technological trends, security becomes the primary concern once the market reaches maturity.

The rise of smart mobile devices has changed the ways that network architecture is managed. Since the introduction of smartphones, more and more user contents have been generated from such mobile devices than client-server computers, and that’s why enterprises have been competing in the offering of cloud services and Big Data management. Thus, the dynamically changing traffic pattern has driven the trends to revolutionize network architectures.

5G is anticipated to greatly benefit IIoT (industrial IoT) with optimized performance, ultra-low latency, and wider network coverage and bandwidth. Among the technological advancements of 5G, network slicing is one of the most expected game changer, enabling cellular operators to divide a physical network architecture into multiple “slices”, which functions as a virtual network, each serving a particular application, cluster or even an ecosystem. More importantly, each virtual network “slice” can be software-defined to meet specific requirements or service, so that QoS (quality of service) and performance are assured with comprehensive, multi-layered security.

Edge computing, or its enhanced version, MEC (multi-access edge computing), has become a proven concept for the foundation of 5G network infrastructure. With the increasing deployments of MEC for the future 5G, the demands for micro data center have skyrocketed, from a US$1.7 billion market in 2015 to become a tremendous US$6.3 billion market by 2020, according to forecasts by MarketsandMarkets.

第 10 頁,共 22 頁