{"title":"High throughput multicast solution for shared memory packet switch","authors":"Kang Xi, N. Ge, C. Feng","doi":"10.1109/HPSR.2002.1024243","DOIUrl":null,"url":null,"abstract":"A high-performance multicast solution is proposed for a shared memory packet switch. In this solution, multicast packets are organized in a single queue and the switching information of the first K packets is processed by a multicast preprocessing arbiter (MPA) to determine the transmission data of each output port. The solution yields high throughput since head-of-line (HOL) blocking can be resolved with a moderate arbiter stage K. Besides, the arbiter also enables packet dropping from the head of the queue in the case of memory overflow, which can eliminate port starvation especially in the case of imbalanced load. This solution can greatly improve multicast throughput without introducing speed and capacity bottlenecks of address memory. The solution can be implemented with moderate complexity. Performance improvement was verified by simulation under balanced and imbalanced load, respectively.","PeriodicalId":180090,"journal":{"name":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2002-11-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"1","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Workshop on High Performance Switching and Routing, Merging Optical and IP Technologie","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/HPSR.2002.1024243","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 1
Abstract
A high-performance multicast solution is proposed for a shared memory packet switch. In this solution, multicast packets are organized in a single queue and the switching information of the first K packets is processed by a multicast preprocessing arbiter (MPA) to determine the transmission data of each output port. The solution yields high throughput since head-of-line (HOL) blocking can be resolved with a moderate arbiter stage K. Besides, the arbiter also enables packet dropping from the head of the queue in the case of memory overflow, which can eliminate port starvation especially in the case of imbalanced load. This solution can greatly improve multicast throughput without introducing speed and capacity bottlenecks of address memory. The solution can be implemented with moderate complexity. Performance improvement was verified by simulation under balanced and imbalanced load, respectively.