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
    • 2. 发明授权
    • Hierarchical output-queued packet-buffering system and method
    • 分层输出排队的分组缓冲系统和方法
    • US06850490B1
    • 2005-02-01
    • US09684763
    • 2000-10-06
    • Leon K. WooRobert Ryan
    • Leon K. WooRobert Ryan
    • H04L12/56H04J1/16
    • H04L49/90H04L47/2441H04L47/50H04L47/6205H04L47/6215H04L49/3036H04L49/9052
    • The packet-buffering system and method of the present invention enables communication devices incorporating a full-mesh architecture to achieve bandwidth aggregation levels ordinarily associated with partial-mesh architectures. The packet-buffering invention uses a hierarchical memory structure having first and second packet-buffers to buffer packets between the input and output ports of the communication device. The received packets are organized by output port and priority level in the first packet buffer, which operates at the aggregate network rate of the communication device. The packets are then funneled to second packet buffers, having corresponding priority and output port assignments, at less than the aggregate network rate and which exhibit buffer depths that exceed that of the first packet buffer. The resulting hierarchical output-queued, packet-buffering system enables a communication system that exhibits a high degree of differentiated services with bandwidth guarantees and at high aggregation levels without experiencing head-of-line blocking.
    • 本发明的分组缓冲系统和方法使得包含全网格架构的通信设备能够实现通常与部分网格架构相关联的带宽聚合级别。 分组缓冲发明使用具有第一和第二分组缓冲器的分层存储器结构来缓冲通信设备的输入和输出端口之间的分组。 接收的分组由第一分组缓冲器中的输出端口和优先级组织,其以通信设备的总网络速率操作。 然后将数据包汇集到具有相对应的优先级和输出端口分配的第二分组缓冲器,小于聚合网络速率,并且其展示的缓冲区深度超过第一分组缓冲器的缓冲区深度。 所产生的分级输出排队的分组缓冲系统使得能够呈现具有带宽保证和高聚合级别的高度差异化服务的通信系统,而不经历线头阻塞。
    • 4. 发明授权
    • Three level distributed control for networking I/O devices
    • 网络I / O设备的三级分布式控制
    • US4905219A
    • 1990-02-27
    • US534562
    • 1983-09-22
    • Paul BarrLeon K. Woo
    • Paul BarrLeon K. Woo
    • H04L12/28
    • H04L12/28
    • A three level network microprocessor based nodal architecture for a fiber optic local area network communication system is disclosed in which the processing responsibility for establishing a voice, video or data communication link between an initiating I/O device and a recipient I/O device and for subsequent two way data exchange between the initiating and recipient devices are shared between three levels of digital processing capability in such manner as to demand the least amount of time from the highest, most intelligent processing level common to all devices thereby making more efficient use of its processing capability as an overall supervisory processor. Each I/O device whether it be a digital terminal, computer or telephone set interfaces with a nodal system through a middle level of processor. An initial communication request from any device at the middle level generates an interrupt command to the highest supervisory processing level which responds to the interrupt, resulting in the identification of the desired recipient station by recalling from storage, associated with the highest level processor, routing information required in establishing a communication link with the desired recipient device. That information is provided to the processor at the middle level associated with the initiating device. That processor completes the link if the recipient is local to it. Otherwise the addresse or recipient device identifying codes are communicated from the middle level to a third level processor responsible for effecting the actual routing of the communication.
    • 公开了一种用于光纤局域网通信系统的基于三电平网络微处理器的节点架构,其中在起始I / O设备和接收者I / O设备之间建立语音,视频或数据通信链路的处理责任, 发起和接收设备之间的后续双向数据交换在数字处理能力的三个级别之间共享,以便从所有设备共同的最高,最智能的处理级别中要求最少的时间量,从而更有效地利用其 作为整体监控处理器的处理能力。 每个I / O设备无论是数字终端,计算机还是电话机通过中间级处理器与节点系统接口。 来自中间级别的任何设备的初始通信请求产生中断命令到最高监控处理级别,该中断命令响应中断,从而通过从与最高级别处理器相关联的存储器调用来确定所需接收方站点的路由信息 与所需的接收设备建立通信链路所需要的。 该信息被提供给处于与发起设备相关联的中间级处理器。 如果收件人是本地的,则处理器完成链接。 否则,地址或接收方设备识别码从中间级传送到负责实现通信的实际路由的三级处理器。