A multifunctional bridge function unit (BRG) and buffer function unit were developed for the OLT LSI for the coexistence of 10G-EPON and GE-PON and implemented in 40-nm CMOS. The units comply with the 10G-EPON standard. The BRG features a high-speed search engine, dual-rate data processing, a low-latency datapath architecture, and a VLAN function. The MAC address search engine enables 22-Gb/s and 33-mega-frame/s throughput by the application of a hash function. The search engine for table look-up enables 11-Gb/s and 16.4-mega-frame/s throughput by the application of eightframe parallel processing and a pipeline configuration. The fullwire rate throughput is achieved for 10-and 1-Gb/s frames simultaneously.
Network microbursts, which are bursts of traffic on the order of submilliseconds, have attracted much attention because they can cause network latency and packet loss. However, analyzing the causes of microbursts involves two problems: how to capture the packets contained in the microbursts and how to identify the flows that cause the microbursts. To solve these problems, we propose a microburst analysis system based on the field-programmable gate array (FPGA). This system uses dedicated hardware to detect microbursts with a submillisecond time resolution. By setting a static threshold for the overall traffic to be detected, the system can capture only packets before and after microburst detection is triggered. In addition, by setting a dynamic threshold for each flow, we can identify the flow that causes the microburst. On the basis of experimental results, we confirmed that our system can accurately identify the flows that cause microbursts by capturing only the packets before and after microburst detection, even in a network where bandwidth usage fluctuates with actual traffic conditions, on the basis of network trace data in a data center. The proposed system is implemented in an Intel PAC with an Arria 10 GX FPGA, which consumes a small amount of hardware resources.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.