Zebra: Building Efficient Network Message Parsers for Embedded Systems

Abstract

Supporting standard text-based protocols in embedded systems is challenging because of the often limited computational resources that embedded systems provide. To overcome this issue, a promising approach is to build parsers directly in the hardware. Unfortunately, developing such parsers is a daunting task for most developers as it is at the crossroads of several areas of expertise, such as low-level network programming or hardware design. In this letter, we propose Zebra, a generative approach that drastically eases the development of hardware parsers and their use in network applications. To validate our approach, we used Zebra to generate hardware parsers for widely used protocols, including HTTP, SMTP, SIP, and RTSP. Our experiments show that Zebra-based parsers are up to 11 times faster than software-based parsers.