A multi-pipeline architecture for high-speed packet classification

AbstractMicrowave holography technique reconstructs a target image using recorded amplitudes and phases of the signals reflected from the target with Fast Fourier Transform (FFT)-based algorithms. The reconstruction algorithms have two or more steps of two- and three-dimensional Fourier transforms, which have a high computational load. In this paper, by neglecting the impact of target depth on image reconstruction, an efficient Fresnel-based algorithm is proposed, involving only one-step FFT for both single- and multi-frequency microwave imaging. Numerous tests have been performed to show the effectiveness of the proposed algorithm including planar and non-planar targets, using the raw data gathered by means of a scanner operating in X-band. Finally, a low-cost and high-speed hardware architecture based on fixed-point arithmetic is introduced which reconstructs the planar targets. This pipeline architecture was tested on field programmable gate arrays operating at 200 MHz clock frequency, which illustrates more than 30 times improvement in computation time compared with a computer.

Download Full-text

AFBV: A High-Performance Network Flow Classification Method for Multi-Dimensional Fields and FPGA Implementation

Journal of Circuits System and Computers ◽

10.1142/s0218126619502372 ◽

2019 ◽

Vol 28 (14) ◽

pp. 1950237

Author(s):

Ling Zheng ◽

Zhiliang Qiu ◽

Weina Wang ◽

Weitao Pan ◽

Shiyong Sun ◽

...

Keyword(s):

High Throughput ◽

Network Flow ◽

High Speed ◽

High Performance ◽

Clock Frequency ◽

Pipeline Architecture ◽

Exact Matching ◽

Flow Classification ◽

Rule Sets ◽

Bit Vector

Network flow classification is a key function in high-speed switches and routers. It directly determines the performance of network devices. With the development of the Internet and various kinds of applications, the flow classification needs to support multi-dimensional fields, large rule sets, and sustain a high throughput. Software-based classification cannot meet the performance requirement as high as 100 Gbps. FPGA-based flow classification methods can achieve a very high throughput. However, the range matching is still challenging. For this, this paper proposes a range supported bit vector (RSBV) method. First, the characteristic of range matching is analyzed, then the rules are pre-encoded and stored in memory. Second, the fields of an input packet header are used as addresses to read the memory, and the result of range matching is derived through pipelined Boolean operations. On this basis, bit vector for any types of fields (AFBV) is further proposed, which supports the flow classification for multi-dimensional fields efficiently, including exact matching, longest prefix matching, range matching, and arbitrary wildcard matching. The proposed methods are implemented in FPGA platform. Through a two-dimensional pipeline architecture, the AFBV can operate at a high clock frequency and can achieve a processing speed of more than 100 Gbps. Simulation results show that for a rule set of 512-bit width and 1[Formula: see text]k rules, the AFBV can achieve a throughput of 520 million packets per second (MPPS). The performance is improved by 44% compared with FSBV and 30% compared with Stride BV. The power consumption is reduced by about 43% compared with TCAM solution.

Download Full-text