会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 1. 发明公开
    • 백본 라우터를 위한 아이피 패킷 주소 룩업 방법 및 그장치
    • 用于在背包路由器的IP包上执行地址查找的方法和装置
    • KR1020040051906A
    • 2004-06-19
    • KR1020020079600
    • 2002-12-13
    • 한국전자통신연구원
    • 진승의곽동용
    • H04L12/28
    • H04L45/54H04L45/745
    • PURPOSE: A method and an apparatus for performing address look-up on an IP packet for a backbone router are provided to maximize packet processing performance by minimizing calculation load of a packet forwarding process. CONSTITUTION: Relay nodes(112b) include position information of child nodes. Termination nodes indicate an entry of a corresponding forwarding information base. A master node(112a) exists at the uppermost position and includes multi-bit search information for the initial uppermost 16 bits. Considering that a network address in an IPv6 global unicast address system consists of upper 64 bits, tree search is performed on the 64 bits. Namely, a multi-bit tree search is first performed on the uppermost 16 bits, and then, the multi-bit tree search is repeatedly performed by extracting every uppermost 8 bits.
    • 目的:提供一种用于对骨干路由器的IP分组进行地址查找的方法和装置,以通过最小化分组转发过程的计算负担来最大化分组处理性能。 构成:中继节点(112b)包括子节点的位置信息。 终止节点指示相应的转发信息库的条目。 主节点(112a)存在于最上位置,并且包括用于初始最高16位的多位搜索信息。 考虑到IPv6全球单播地址系统中的网络地址由高64位组成,64位执行树搜索。 即,首先在最上面的16位执行多位树搜索,然后通过提取每个最高8位重复执行多位树搜索。
    • 2. 发明公开
    • 차등화된 QoS 서비스를 제공하는 라우터 및 그것의고속 IP 패킷 분류 방법
    • 提供差分QOS服务的路由器和高速分类IP分组的方法
    • KR1020030089747A
    • 2003-11-28
    • KR1020020027594
    • 2002-05-18
    • 한국전자통신연구원
    • 진승의김태일이형호
    • H04L12/28
    • H04L47/10H04L45/00H04L45/302H04L45/60H04L45/742H04L47/2408H04L47/2441H04L49/90
    • PURPOSE: A router for providing differentiated QoS(Quality of Service) service and a method for classifying IP packets at a high speed are provided to maximize the performance of packet classification by minimizing the operation loads generated due to packet classification in the process of routing. CONSTITUTION: A packet classifier(60) is composed of a search path part(610) and a control part(690). The search path part(610) determines relevant flow processing by searching flow tables for each inputted packet, and comprises a multi-field search key extraction part(620), a differentiated class classifier(630), differentiated class flow table searchers(641-64m), an on-chip memory(660) containing differentiated class flow tables(661-66m), and differentiated class off-chip memory interfaces(681-68m). The multi-field search key extraction part(620) extracts multi-field information from an inputted packet and creates a multi-field search key. The differentiated class classifier(630) executes differentiated class classification by an exact matching method, based on the value of a differentiated service code point field contained in the created multi-field search key. The differentiated class flow table searchers(641-64m) execute searching for associated flow tables(661-66m) in parallel according to classified differentiated classes. In case that flow table searching at the on-chip memory(660) is not successful, the differentiated class off-chip memory interfaces(681-68m) receive search requests from differentiated class flow table searchers(641-64m) and search the flow tables constructed at an off-chip memory.
    • 目的:提供用于提供差异化​​QoS(Quality of Service)服务的路由器和用于高速分类IP分组的方法,以通过最小化在路由过程中由于分组分类而产生的操作负载来最大化分组分类的性能。 构成:分组分类器(60)由搜索路径部分(610)和控制部分(690)组成。 搜索路径部分(610)通过搜索每个输入的分组的流程表来确定相关的流程处理,并且包括多字段搜索关键词提取部分(620),差分分类器(630),差分类流程表搜索器(641- 64m),包含差分类流表(661-66m)的片上存储器(660)和差分类外片外存储器接口(681-68m)。 多字段搜索关键词提取部分(620)从输入的分组提取多字段信息并创建多字段搜索关键字。 差分类分类器(630)基于所创建的多字段搜索关键字中包含的差分服务码点字段的值,通过精确匹配方法执行差分类分类。 差分类流表搜索器(641-64m)根据分类的分类进行并行搜索相关流表(661-66m)。 在片上存储器(660)的流表搜索不成功的情况下,差分类外片存储器接口(681-68m)接收来自差分类流表搜索器(641-64m)的搜索请求并搜索流 在片外存储器中构造的表。