公开(公告)号：US06483854B1

公开(公告)日：2002-11-19

申请号：US09774335

申请日：2001-01-30

申请人： Daniel E. Klausmeier ,  Kevin Wong ,  David A. Hughes

发明人： Daniel E. Klausmeier ,  Kevin Wong ,  David A. Hughes

IPC分类号： H04J324

CPC分类号： H04L49/9094 ,  H04L49/108 ,  H04L49/90 ,  H04L49/9021 ,  H04L2012/5647 ,  H04L2012/5658 ,  H04L2012/5665

摘要： A plurality of cells are received from a digital network and stored in a logical queue. The cells contain data information and logical connection information and the logical queue corresponds to the logical connection identified by the logical connection information. The cells are combined into a frame by extracting the cells from the logical queue and storing the cells to a local memory so that the data information is preserved. Extraction may be accomplished by notifying a local processor that the frame's worth of data is ready and then transferring the data information of the cells to a local memory at the direction of the local processor. During the transfer, error detection operations may be performed. Information may be transmitted into the digital network by segmenting a frame of data into a plurality of cells and injecting each of the cells into a logical queue. The logical queue may be constructed with a series of linked list pointers associated with the memory locations. The cells may then be transmitted into the digital network in an order consistent with the linked list pointers.

摘要翻译： 多个小区从数字网络接收并存储在逻辑队列中。 小区包含数据信息和逻辑连接信息，逻辑队列对应于由逻辑连接信息标识的逻辑连接。 通过从逻辑队列提取单元格并将单元格存储到本地存储器中，将单元格组合成一帧，从而保留数据信息。 提取可以通过通知本地处理器该帧的数据准备就绪，然后在本地处理器的方向上将单元的数据信息传送到本地存储器来完成。 在传送期间，可以执行错误检测操作。 信息可以通过将数据帧分割成多个小区并将每个小区注入到逻辑队列中来发送到数字网络。 可以使用与存储器位置相关联的一系列链接列表指针来构造逻辑队列。 然后可以以与链表指针一致的顺序将单元发送到数字网络。

公开(公告)号：US06430191B1

公开(公告)日：2002-08-06

申请号：US08885400

申请日：1997-06-30

申请人： Daniel E. Klausmeier ,  Kevin Wong ,  Quang Nguyen ,  Cherng-Ren Sue ,  David A. Hughes ,  Ross Suydam Heitkamp ,  Rafael Gomez

发明人： Daniel E. Klausmeier ,  Kevin Wong ,  Quang Nguyen ,  Cherng-Ren Sue ,  David A. Hughes ,  Ross Suydam Heitkamp ,  Rafael Gomez

IPC分类号： H04L1256

CPC分类号： H04L47/10 ,  H04L12/5601 ,  H04L47/29 ,  H04L47/33 ,  H04L49/3081 ,  H04L2012/5636 ,  H04L2012/5648 ,  H04L2012/5651 ,  H04L2012/5679 ,  H04L2012/5681 ,  H04L2012/5682

摘要： In a digital switch, incoming cells are placed into a queue in a cell memory. The switch maintains various cell queues, including VC queues that correspond to individual connections and QBin queues that correspond to various classes of service. Cells may arrive to a VC queue or a QBin queue but will depart from a QBin queue. Accordingly, cells may be moved from VC queues to QBin queues. Cells are serviced according to the use of QBin Groups. A QBin Group (QBG) includes a number of logical queues (QBins) of cells to be transported in the digital network. After a QBG is selected, one of its logical queues is selected for servicing. The QBG may be selected by examining all of the QBGs to find an eligible QBG which is most overdue for service. A QBin of the selected QBG may then be selected by examining each of the QBins comprising the selected QBG to find the most overdue for service. The QBGs may correspond to virtual interfaces.

摘要翻译： 在数字交换机中，传入的信元被放入单元存储器中的队列中。 交换机维护各种小区队列，包括对应于各种连接的VC队列和对应于各种服务类别的QBin队列。 单元格可能会到达一个VC队列或一个QBin队列，但会离开QBin队列。 因此，小区可以从VC队列移动到QBin队列。 根据QBin组的使用对单元进行维修。 QBG组（QBG）包括在数字网络中要传输的单元的多个逻辑队列（QBins）。 选择QBG后，选择其逻辑队列之一进行维修。 QBG可以通过检查所有QBG来选择，以找到最适合服务的合格QBG。 然后可以通过检查包括所选择的QBG的每个QB，来查找所选择的QBG的QBin，以找到服务最迟。 QBG可以对应于虚拟接口。

公开(公告)号：US06487202B1

公开(公告)日：2002-11-26

申请号：US08884999

申请日：1997-06-30

申请人： Daniel E. Klausmeier ,  Kevin Wong

发明人： Daniel E. Klausmeier ,  Kevin Wong

IPC分类号： H04L1228

CPC分类号： H04L12/5601 ,  H04L47/24 ,  H04L49/108 ,  H04L49/90 ,  H04L49/901 ,  H04L49/9089 ,  H04L49/9094 ,  H04L2012/5679 ,  H04L2012/5681

摘要： A method of executing a sequence of multiple dependent operations, each operation including a memory read and a memory write involves overlapping memory accesses of the operations by grouping together memory reads and memory writes of multiple operations and preserving a desired sequence of the operations using a circuit external to a memory through which the memory accesses are performed. The operations may be updates to one or more linked lists. In one embodiment, the step of overlapping memory accesses may be performed by grouping together memory accesses according to ATM cell arrivals or departures. In this embodiment, the operations are associated with ATM cell arrivals or departures and may be gets or puts. Each get and put operation may be characterized by a number of atomic memory operations to update one or more linked lists. To perform the operations a circuit a having an address processor, a data processor coupled to the address processor and to the external memory, and a prefetch buffer coupled to the external memory, the address processor and to the data processor is provided. The address processor generates memory addresses for the operations according to the step of overlapping memory accesses. The atomic memory operations are grouped so that all of the memory read operations associated with the dependent operations are performed before all of the memory write operations associated with the dependent operations are performed.

摘要翻译： 一种执行多个相关操作的序列的方法，每个包括存储器读取和存储器写入的操作都涉及通过将存储器读取和多个操作的存储器写入分组在一起而使操作的存储器访问重叠，并使用电路 执行存储器访问的存储器外部。 这些操作可以是对一个或多个链表的更新。 在一个实施例中，可以通过根据ATM信元到达或离开将存储器访问分组在一起来执行重叠存储器访问的步骤。 在该实施例中，操作与ATM信元到达或离开相关联，并且可以被获取或放置。 每个获取和放置操作可以通过多个原子存储器操作来表征，以更新一个或多个链表。 为了执行具有地址处理器，耦合到地址处理器和外部存储器的数据处理器以及耦合到外部存储器的预取缓冲器的地址处理器和数据处理器的电路。 地址处理器根据重叠存储器访问的步骤产生用于操作的存储器地址。 原子存储器操作被分组，使得在执行与依赖操作相关联的所有存储器写入操作之前执行与从属操作相关联的所有存储器读取操作。

公开(公告)号：US06201813B1

公开(公告)日：2001-03-13

申请号：US08884705

申请日：1997-06-30

申请人： Daniel E. Klausmeier ,  Kevin Wong ,  David A. Hughes

发明人： Daniel E. Klausmeier ,  Kevin Wong ,  David A. Hughes

IPC分类号： H04L1228

CPC分类号： H04L49/9094 ,  H04L49/108 ,  H04L49/90 ,  H04L49/9021 ,  H04L2012/5647 ,  H04L2012/5658 ,  H04L2012/5665

摘要： A plurality of cells are received from a digital network and stored in a logical queue. The cells contain data information and logical connection information and the logical queue corresponds to the logical connection identified by the logical connection information. The cells are combined into a frame by extracting the cells from the logical queue and storing the cells to a local memory so that the data information is preserved. Extraction may be accomplished by notifying a local processor that the frame's worth of data is ready and then transferring the data information of the cells to a local memory at the direction of the local processor. During the transfer, error detection operations may be performed. Information may be transmitted into the digital network by segmenting a frame of data into a plurality of cells and injecting each of the cells into a logical queue. The logical queue may be constructed with a series of linked list pointers associated with the memory locations. The cells may then be transmitted into the digital network in an order consistent with the linked list pointers.

摘要翻译： 多个小区从数字网络接收并存储在逻辑队列中。 小区包含数据信息和逻辑连接信息，逻辑队列对应于由逻辑连接信息标识的逻辑连接。 通过从逻辑队列提取单元格并将单元格存储到本地存储器中，将单元格组合成一帧，从而保留数据信息。 提取可以通过通知本地处理器该帧的数据准备就绪，然后在本地处理器的方向上将单元的数据信息传送到本地存储器来完成。 在传送期间，可以执行错误检测操作。 信息可以通过将数据帧分割成多个小区并将每个小区注入到逻辑队列中来发送到数字网络。 可以使用与存储器位置相关联的一系列链接列表指针来构造逻辑队列。 然后可以以与链表指针一致的顺序将单元发送到数字网络。

公开(公告)号：US07672301B2

公开(公告)日：2010-03-02

申请号：US10428048

申请日：2003-05-02

申请人： Daniel E. Klausmeier ,  Edward Sprague

发明人： Daniel E. Klausmeier ,  Edward Sprague

IPC分类号： H04Q11/00

CPC分类号： H04Q3/68

摘要： A distribution stage is disclosed comprising a plurality of inputs coupled to a plurality of first stage switching devices, a plurality of outputs coupled to a plurality of second stage switching devices, and a distribution configuration. The distribution configuration is configured to receive a plurality of bandwidth units (BU's) from each first stage switching device, and to distribute at least one BU from each first stage switching device to each second stage switching device, such that each second stage switching device is assured of receiving at least one BU from each first stage switching device. In effect, the distribution stage ensures that each first stage switching device has a logical link to each second stage switching device. In one embodiment, the distribution stage is configured in accordance with a distribution configuration that is static. Because the distribution configuration of the distribution stage is static, it does not add complexity to the overall switching determination of a switching network. Thus, the distribution stage enables the switching network to be expanded without degrading the switching performance of the network.

摘要翻译： 公开了一种分配阶段，包括耦合到多个第一级交换设备的多个输入，耦合到多个第二级交换设备的多个输出以及分配配置。 分配配置被配置为从每个第一级交换设备接收多个带宽单元（BU），并且将来自每个第一级交换设备的至少一个BU分配给每个第二级交换设备，使得每个第二级交换设备是 确保从每个第一级切换装置接收至少一个BU。 实际上，分配阶段确保每个第一级交换设备具有到每个第二级交换设备的逻辑链路。 在一个实施例中，分配阶段根据静态的分布配置来配置。 由于分布阶段的分布配置是静态的，因此交换网络的整体交换确定不会增加复杂性。 因此，分发阶段使得能够扩展交换网络，而不降低网络的交换性能。

公开(公告)号：US5666353A

公开(公告)日：1997-09-09

申请号：US408273

申请日：1995-03-21

申请人： Daniel E. Klausmeier ,  Charles M. Corbalis ,  Kambiz Hooshmand

发明人： Daniel E. Klausmeier ,  Charles M. Corbalis ,  Kambiz Hooshmand

IPC分类号： H04L12/56 ,  H04Q11/04

CPC分类号： H04L12/5602 ,  H04Q11/0478 ,  H04L2012/5636 ,  H04L2012/5637 ,  H04L2012/5648 ,  H04L2012/5652

摘要： A frame based traffic policing system that determines if incoming data cells are conforming or non-conforming according to the a traffic contract. The frame based traffic policing system first detects a cell at the beginning or end of a frame and determines if the frame conforms with a traffic contract. If the cell does not exceed the parameters of the traffic contract, then the frame based traffic policing system deems the cell as conforming. If the cell was the first cell of a frame, then frame based traffic policing system treats all the remaining cells in the frame as conforming or non-conforming depending upon if the first data cell was conforming or non-conforming. If the cell was the last cell of a frame, then frame based traffic policing system treats all the cells of the following frame as conforming or non-conforming depending upon if the last cell of a previous frame was conforming or non-conforming. In this manner, all the data cells are from the same frame receive the same treatment by the network. The frame based traffic policing system improves the throughout of the network since the network will be less likely to waste resources transmitting cells from a corrupted frame.

摘要翻译： 基于帧的流量监管系统，其根据流量合同确定输入数据单元是否符合或不符合。 基于帧的流量监管系统首先检测帧的开始或结束处的小区，并确定帧是否符合交通契约。 如果小区不超过业务合约的参数，则基于帧的流量监管系统认为小区符合要求。 如果小区是帧的第一个小区，则基于帧的流量监管系统将帧中的所有剩余小区视为一致或不一致，这取决于第一个数据单元是否符合或不一致。 如果小区是帧的最后一个小区，则基于帧的流量监管系统将根据前一帧的最后一个小区是否符合或不一致，将以下帧的所有小区视为一致或不一致。 以这种方式，来自同一帧的所有数据信元都由网络接收相同的处理。 基于帧的流量监管系统改善了整个网络，因为网络将不太可能浪费从损坏的帧传输小区的资源。

公开(公告)号：US5835494A

公开(公告)日：1998-11-10

申请号：US825409

申请日：1997-03-27

申请人： David A. Hughes ,  Daniel E. Klausmeier

发明人： David A. Hughes ,  Daniel E. Klausmeier

IPC分类号： H04L12/56 ,  H04Q11/04

CPC分类号： H04L12/5602 ,  H04L12/5693 ,  H04L47/2441 ,  H04L47/568 ,  H04Q11/0478 ,  H04L2012/5649 ,  H04L2012/5679 ,  H04L2012/5681

摘要： A method and apparatus are provided for scheduling when each of a plurality of virtual connections supported by a transmit device will be serviced by the transmit device. The transmit device includes a transmission control unit that uses a plurality of calendars to schedule when each of the plurality of virtual connections will be serviced. Virtual connections with faster transfer rates are scheduled using higher granularity calendars, while virtual connections with slower transfer rates are scheduled using lower granularity calendars. Each entry in the calendars is associated with a time interval ("cell slot") during which the transmit device is able to service one virtual connection. During a given cell slot, linked lists associated with the calendar entries that correspond to the cell slot are added to a service queue, and the virtual connection at the head of the service queue is serviced. After a virtual connection is serviced, the next service time for the virtual connection is scheduled by attaching the virtual connection to the head of a linked list associated with the appropriate entry of the appropriate calendar. The appropriate entry is determined by incrementing a fixed-point "desired service time" value that represents the cell slot for which the virtual connection was last scheduled, by a fixed-point inter-gap value that represents the how many cell slots ideally elapse between sequential service times of the virtual connection.

摘要翻译： 提供了一种方法和装置，用于当传输设备支持的多个虚拟连接中的每一个将由发送设备服务时进行调度。 发送设备包括传输控制单元，其使用多个日历来调度何时将维护多个虚拟连接中的每一个。 使用更高粒度的日历调度具有更快传输速率的虚拟连接，而使用较低粒度的日历安排使用较慢传输速率的虚拟连接。 日历中的每个条目与发送设备能够服务于一个虚拟连接的时间间隔（“小区时隙”）相关联。 在给定的小区时隙期间，与对应于小区时隙的日历条目相关联的链接列表被添加到服务队列中，并且服务队列头部处的虚拟连接被服务。 在虚拟连接被服务之后，通过将虚拟连接附加到与适当日历的适当条目相关联的链接列表的头部来安排虚拟连接的下一服务时间。 通过将表示虚拟连接最后调度的小区时隙的定点“期望服务时间”值增加一个定点间隙值来确定适当的条目，该定点间隙值表示理想地经过多少小区时隙 虚拟连接的连续服务时间。

公开(公告)号：US5570360A

公开(公告)日：1996-10-29

申请号：US407906

申请日：1995-03-20

申请人： Daniel E. Klausmeier ,  Charles M. Corbalis

发明人： Daniel E. Klausmeier ,  Charles M. Corbalis

IPC分类号： H04L12/56 ,  H04Q11/04 ,  H04L12/54

CPC分类号： H04L12/5602 ,  H04Q11/0478 ,  H04L2012/5615 ,  H04L2012/5635 ,  H04L2012/5636 ,  H04L2012/5637 ,  H04L2012/5638 ,  H04L2012/5679 ,  H04L2012/568 ,  H04L2012/5682

摘要： To simplify the service decision made in a queue based digital switch, a method and apparatus is introduced that makes service decisions based upon the history of incoming cells. The apparatus includes a minimum service rate circuit that monitors the rate of arriving cells. The minimum service rate circuit tests the arriving cells to determine if they are in conformance with the minimum service rate of a service contract. The minimum service rate circuit may use a continuous state leaky bucket algorithm as specified in the CCITT Recommendation I.371 document, dated 1992, to determine if a cell is in conformance with the minimum service rate of the service contract. When a cell arrives that conforms to the minimum service rate, the minimum service rate circuit notifies the server and the server serves the queue associated with the connection.

摘要翻译： 为了简化基于队列的数字交换机的服务决策，引入了一种基于传入单元的历史进行服务决策的方法和装置。 该装置包括监视到达单元的速率的最小服务速率电路。 最小服务速率电路测试到达的信元，以确定它们是否符合服务合同的最低服务速率。 最小服务速率电路可以使用1992年CCITT建议I.371文件中规定的连续状态泄漏桶算法，以确定一个小区是否符合服务合同的最低服务速率。 当小区到达符合最低服务速率时，最小服务速率电路通知服务器，服务器服务与该连接相关联的队列。

公开(公告)号：US07184432B2

公开(公告)日：2007-02-27

申请号：US10704976

申请日：2003-11-12

申请人： Joel F. Adam ,  Darren Engelkemier ,  Daniel E. Klausmeier

发明人： Joel F. Adam ,  Darren Engelkemier ,  Daniel E. Klausmeier

IPC分类号： H04J3/24 ,  H04L12/50

CPC分类号： H04L45/00 ,  H04J2203/0012 ,  H04J2203/0023 ,  H04J2203/006 ,  H04L45/54 ,  Y10T307/76

摘要： A switch is provided that receives user information through a plurality of framer circuits, which group the user information into frames. The frames are fed to a switch fabric including an array of switch elements, each having a switch matrix for routing each frame to a desired output in accordance with configuration data stored in a first table coupled to the switch matrix. If different outputs are desired, i.e., the switch matrix is to be reconfigured, a switch control circuit supplies additional switch configuration data to the frames through the inputs along with additional user information to be routed through the switch. While the additional switch configuration data is stored in a second table, data flow remains uninterrupted through the switch matrix. Once storage of the additional configuration data into the second table is complete, however, the switch control circuit inserts a table select signal into the frames, to thereby couple the second table to the switch so that the switch matrix is configured in accordance with the additional switch configuration data. Subsequent frames are then routed through the reconfigured switch matrix. The first and second tables can thus be alternately updated and coupled to the switch matrix to appropriately reconfigure the switch without affecting existing connections established through the switch. “Hitless switchover” is therefore achieved.

摘要翻译： 提供了一种通过多个成帧器电路接收用户信息的开关，其将用户信息分组成帧。 帧被馈送到包括开关元件阵列的交换结构，每个开关元件具有开关矩阵，用于根据存储在耦合到开关矩阵的第一表中的配置数据将每个帧路由到期望的输出。 如果需要不同的输出，即开关矩阵将被重新配置，则开关控制电路通过输入将附加的开关配置数据与附加的用户信息一起提供给要通过开关路由的附加的用户信息。 当附加的开关配置数据存储在第二个表中时，数据流通过开关矩阵保持不间断。 然而，一旦将另外的配置数据存储到第二表中，则开关控制电路将表选择信号插入到帧中，从而将第二表耦合到开关，使得开关矩阵根据额外的 切换配置数据。 随后的帧然后通过重新配置的交换矩阵路由。 因此，第一和第二表可以交替更新并耦合到开关矩阵，以适当地重新配置开关，而不会影响通过开关建立的现有连接。 因此实现“无意义切换”。

公开(公告)号：US07020135B2

公开(公告)日：2006-03-28

申请号：US10023971

申请日：2001-12-17

申请人： Daniel E. Klausmeier ,  Jeffrey T. Gullicksen

发明人： Daniel E. Klausmeier ,  Jeffrey T. Gullicksen

IPC分类号： H04L12/50

CPC分类号： H04Q3/68 ,  H04Q2213/1302 ,  H04Q2213/1304 ,  H04Q2213/13046 ,  H04Q2213/13166 ,  H04Q2213/13292 ,  H04Q2213/1334

摘要： A switch is provided that includes three stages. The first stages has a plurality of switch circuits. The second stage has a plurality of switch circuits equal to N, where N is any integer other than a power of 2 and where the switch circuits can be logically configured into a logical configuration of a power of 2. The third stages includes a plurality of switch circuits.