The rise of network speeds to tens of gigabits per second poses a challenge to develop packet processing applications that can cope with such bit rates. Therefore, the need for a suitable open source system that can be used as a prototype platform to test new network functionality while ensuring line-rate processing, accurate timestamping, and reduced power consumption has become evident. All these requirements cannot be achieved by using software-only solutions, but rather with hardware-based platforms such as NetFPGA. The main obstacle when using this type of open source FPGA-based solution is the cost of development, in both time and hardware development skills required. The spread of new circuit synthesis tools using high-level languages opens the door for the development of hardware-based networking applications with reasonable development effort, compared to the use of traditional hardware description languages. In this article, we describe how existing open source hardware-based platforms for networking applications will be fueled by the change in the programming model of FPGAs provided by modern high-level synthesis tools. For this, we implemented a network flow monitor using high-level languages and compared the effort spent with respect to a traditional hardware development cycle. Preliminary results are very promising, given that development time is reduced from months to weeks.
[1]
Michaela Blott.
Fpga Research Design Platform Fuels Network Advances Fpga Research Design Platform Fuels Network Advances
,
.
[2]
Gustavo Sutter,et al.
Accurate and flexible flow-based monitoring for high-speed networks
,
2013,
2013 23rd International Conference on Field programmable Logic and Applications.
[3]
José Luis García-Dorado,et al.
Batch to the Future: Analyzing Timestamp Accuracy of High-Performance Packet I/O Engines
,
2012,
IEEE Communications Letters.
[4]
Aiko Pras,et al.
On the future of Internet management technologies
,
2003,
IEEE Commun. Mag..
[5]
Dominique Lavenier,et al.
HLS Tools for FPGA: Faster Development with Better Performance
,
2011,
ARC.
[6]
George Varghese,et al.
Building a better NetFlow
,
2004,
SIGCOMM.
[7]
Gary Smith,et al.
High-Level Synthesis: Past, Present, and Future
,
2009,
IEEE Design & Test of Computers.
[8]
Jean-Pierre Deschamps,et al.
Guide to FPGA Implementation of Arithmetic Functions
,
2012,
Lecture Notes in Electrical Engineering.