Ethernity Networks and Advantech announced that they have successfully integrated an effective virtual Broadband Remote Access Server (vBRAS) platform that offers outstanding performance at the network edge, maximizing throughput without impacting CPU efficiency.
The joint solution utilizes high-performance Advantech SKY-8000 series servers designed for carrier-grade deployment, and featuring Intel Xeon Scalable processors, to provide a vBRAS platform with a scalable data plane and independent software-based control plane.
The servers are equipped with Ethernity ACE-NIC smart network adaptors, which not only provide optimized data transfer using patented ENET Flow Processor technology, but also offload resource-consuming virtual functions to FPGAs, accelerating performance and providing the necessary deterministic results and low latency for a carrier-grade vBRAS solution. In turn, this reduces total cost of ownership by enabling accelerated performance.
As the main gateway element to the internet, BRAS platforms must support massive throughput and be scalable enough to support the significant growth in data usage that is expected from ultrabroadband services. The joint Advantech-Ethernity vBRAS solution addresses the scalability challenge, enables reduced power and space requirements, and frees the CPU to run additional user sessions, resulting in lower latency and deterministic performance. Moreover, the programmable nature of the FPGA-based platform provides a truly open solution with complete serviceability.
Eugene Zetserov, VP of Products, Ethernity Networks
By pairing Advantech servers with Ethernity’s data plane over FPGA and control plane software, we propose a total solution as a value added to traditional servers.
James Yang, VP Advantech Networks & Communications Group
Ethernity’s FPGA technology adds an advanced network acceleration element to Advantech’s SKY Server line, enabling service providers with the performance levels they need to viably migrate from today’s closed proprietary designs. Fully supported by a comprehensive x86 software framework, integrators and service providers alike can decide when best to turn on hybrid acceleration assets to meet required cost per packet thresholds.
