April 6, 2017 — Netcope Technologies, a company dedicated to development and distribution of FPGA solutions for the fastest networks has recently achieved 148.8 Mega packets per second (Mpps) on Data Plane Development Kit (DPDK), an open source API designed by Intel for optimization of data transfer from network interface cards to Intel CPUs. Not only is 148.8 the highest throughput achieved by any solution provider, but it is also the highest theoretical throughput, since this is throughput fastest optical links of today have (100G). This has been achieved with a layered solution combining Netcope Data Plane API, NFB-100G2Q Netcope network interface card and DPDK API. Achieving maximum theoretical throughput is good news for any communication service provider who wants to enjoy efficient deployment of Network Function Virtualization (NFV) applications, lowering the amount of machines required for Virtual Network Functions (VNFs).
Since many commercial and open source applications support DPDK and it became a de facto standard in the NFV world. Open vSwitch, FD.IO VPP, or TRex traffic generator are widely deployed examples of such
applications. The core principle of how NFV functions is that there is a commodity server that runs the virtual function, just like its dedicated hardware counterpart would. The problem on fastest networks of today (100 Gbps) is that the commodity server can hardly achieve the level of performance dedicated hardware would. This presents a market opportunity for vendors who are capable of developing appliances for hardware acceleration. Netcope responds with FPGA-based network interface cards.
The network interface card processes the traffic inside its FPGA chip and sends the traffic to the CPU. DPDK here helps by improving throughput between the card and the CPU. The combination of DPDK and NFB-100G2Q reaches the desired throughput of 148.8 Mpps only at certain packet length, because of the overhead in traffic on the shortest packets:
Netcope has recently developed Netcope Data Plane API that communicates well with DPDK and according to the latest tests, the combination of NDP, NFB-100G2Q and DPDK has no problem achieving maximum
theoretical throughput of 148,8 Mpps even on the shortest packets:
The main added value of the record breaking throughput guaranteed by DPDK accelerated by NDP is that fewer commodity servers are required to process computation intensive VNFs, because more packets are forwarded to CPU at any given moment, so VNFs can be executed faster. And fewer machines required for the same amount of tasks means less power required, rack space saved, downtime and failure expenses reduced, which all sums up in reduced total cost of ownership of the whole operation.
Netcope Technologies is a leading manufacturer and provider of high-performance network solutions. We excel in packet capture and packet processing technologies and low-latency trading solutions. Our focus is on delivering state-of-the-art solutions for high-speed and low-latency networks. Our products are deployed world-wide. The Netcope Technologies portfolio covers the whole field of products for the hardware acceleration of network traffic processing using FPGA technology. These products are ideal for all OEM vendors, R&D customers and end customers to build, develop and deploy hardware-accelerated solutions. The use of our network adapters with advanced features enables our customers and partners to gain a competitive advantage in the world of high-speed and low-latency networks.
Source: Netcope Technologies