Norio Yamagaki

High-speed regular expression matching engine using multi-character NFA

An approach is presented for high throughput matching of regular expressions (regexes) by first converting them into corresponding non-deterministic finite automata (NFAs) which are then configured onto a FPGA. The key novel feature is a technique that, for any given regex, constructs an NFA that processes multiple characters per clock cycle. An efficient algorithm is proposed that outputs an NFA which processes twice the number of characters as the input one. A technique is also proposed that implements the range match operation (e.g. [a-z]) efficiently. A program has been written that implements above ideas to convert regexes into NFAs specified in a structural hardware design language (HDL), which are then mapped onto a FPGA. Performance is evaluated using real world regexes (Snort ruleset). The results demonstrate the practical utility of the approach. For example, for a set of 2,691 regexes, while the standard 1-character NFA obtains a throughput of 1.25 Gbps, our 4-character NFA achieves a throughput of 3.63 Gbps, while requiring only 20% more LUTs and 6% less flip-flops.

Decision Forest: A Scalable Architecture for Flexible Flow Matching on FPGA

Next generation Internet requires processing rich and flexible flow information in the network infrastructure. Rapid growth in network traffic results in major challenge to support flexible flow matching at line rate. Most of the existing work focuses on functionality rather than performance, and simply adopts either power-hungry TCAM or performance-in deterministic hashing. This paper exploits the abundant parallelism and other desirable features provided by state-of-the-art FPGAs, and proposes a parallel architecture, named decision forest, for high-performance flexible flow matching. We develop a framework to partition a given table of flexible flow rules into multiple subsets each of which is built into a depth-bounded decision tree. The partitioning scheme is carefully designed to reduce rule duplication during the construction of the decision trees. Thus the overall memory requirement is significantly reduced. After such partitioning, the number of header fields used to build the decision tree for each rule subset is small. This leads to reduction in logic resource requirement. Exploiting the dual-port RAMs available in current FPGAs, we map each decision tree onto a linear pipeline to achieve high throughput. Our extensive experiments and FPGA implementation demonstrate the effectiveness of our scheme. Our design supports 1K flexible flow rules while sustaining 40 Gbps throughput for matching minimum size (40 bytes) packets. To the best of our knowledge, this is the first FPGA design for flexible flow matching to achieve over 10 Gbps.

An efficient one-to-many content distribution method for Wi-Fi networks

Recently, along with the widespread use of Wi-Fi devices, there has been high expectation for contents sharing in the Wi-Fi networks that consists of large number of nodes. In this paper, we present an efficient one-to-many content distribution method for Wi-Fi networks. In our method, a source node selects a representative node from destination nodes which requested for the same content and then sends the content to the selected representative node by using TCP. At the same time, the rest of destination nodes receive the content by overhearing TCP traffic generated from the source node to the representative node. We also present a technique to enable overhearing in an environment which employs pairwise key and group key based encryption mechanism such as WPA. Finally, we evaluate the effectiveness of our method through simulations using QualNet. Our simulation results show that our method can shorten the content delivery time up to 94% compared with content distribution using FTP in the network consists of 500 nodes.

Implementation and evaluation of efficient one-to-many content distribution method in wireless LAN with high node density

Along with the rapid increasing use of wireless LAN (Wi-Fi) enabled devices, nowadays it is becoming possible for users to exchange data wirelessly without the need of network infrastructure. Therefore, during disaster time when network infrastructure is not available, utilizing Wi-Fi as a means for information sharing to those in the disaster affected area can be a promising solution. Especially, in such situations delivering information to several hundred of victims at the refuge site or stranded commuters at stations is important. However, to make this possible, fast content distribution method in large scale Wi-Fi network with high node density is required. To achieve this goal, we have proposed an efficient one-to-many content distribution method. In this paper, we show the evaluation results of our proposed method in wireless LAN with high node density through simulations and experiments. Through simulations, we confirmed that our approach outperforms content distribution using FTP in wireless LAN consists of several hundred nodes. Furthermore, we also implemented a prototype of our method as a software. Through experiments by installing our software on tablets, we confirmed the feasibility of our method in real Wi-Fi network which consists of 100 nodes and found that it is expectable that our method will outperform the conventional method as the number of nodes increases.