公开(公告)号：US07085275B2

公开(公告)日：2006-08-01

申请号：US10131467

申请日：2002-04-25

申请人： Jen-Kai Chen ,  Yun-Fei Chao

发明人： Jen-Kai Chen ,  Yun-Fei Chao

IPC分类号： H04L12/56 ,  G06F12/00

CPC分类号： G06F17/30949 ,  H04L29/12801 ,  H04L29/12839 ,  H04L45/745 ,  H04L61/6004 ,  H04L61/6022

摘要： A method and apparatus for accessing a non-symmetric dual-slot address table having two entries 0, 1 that are different in size. When writing data, the data is hashed to generate a hash value. Then, the data is written to a slot corresponding to the hash value in the entry 0. When there was data in the slot 0, the hash value is mapped to a sub-address and a share value, and then a slot 1 corresponding to the sub-address is selected from the entry 1. Afterwards, the SMAC tag, the share value and the source port are written to the slot 1. In addition, when transmitting the packet, a hash value is generated according to the DMAC address. Then, a SMAC tag, the share value and the source port are read according to the hash value. After compared, the packet is transmitted.

公开(公告)号：US07092387B2

公开(公告)日：2006-08-15

申请号：US09821428

申请日：2001-03-29

申请人： Jen-Kai Chen ,  Chao-Cheng Cheng ,  Jiann-Hwa Liou

发明人： Jen-Kai Chen ,  Chao-Cheng Cheng ,  Jiann-Hwa Liou

IPC分类号： H04L12/50

CPC分类号： H04L12/5602 ,  H04L47/13 ,  H04L47/32

摘要： A method for easing data transmission congestion in a switch device having a plurality of ports. The switch device includes a shared buffer capable of dividing into a plurality of buffering units. The method comprises the steps as follows. A plurality of output queues are provided, in which the output queues have a one-to-one correspondence with the ports, and each output queue has a number of reserved buffering units. The buffering unit(s) are then assigned in the shared buffer. A packet is received from a source port and storing the packet in an assigned buffering unit. The target port of a packet is determined according to a header of the packet. Then, the buffering unit containing the packet is linked to the output queue that corresponds to the target port. The free buffering units are thus controlled according to a number of reserved buffering units in the output queue and a triggering or a terminating condition of the source port.

摘要翻译： 一种用于在具有多个端口的交换设备中缓解数据传输拥塞的方法。 开关装置包括能够分成多个缓冲单元的共享缓冲器。 该方法包括以下步骤。 提供多个输出队列，其中输出队列与端口具有一一对应关系，并且每个输出队列具有多个保留的缓冲单元。 然后将缓冲单元分配在共享缓冲器中。 从源端口接收分组并将分组存储在分配的缓冲单元中。 分组的目标端口根据分组的报头来确定。 然后，包含分组的缓冲单元被链接到对应于目标端口的输出队列。 因此，可以根据输出队列中的保留缓冲单元数量和源端口的触发或终止条件来控制自由缓冲单元。

公开(公告)号：US06958973B1

公开(公告)日：2005-10-25

申请号：US09669879

申请日：2000-09-27

申请人： Jen-Kai Chen ,  Yu-Ju Lin ,  Jiann-Hwa Liou

发明人： Jen-Kai Chen ,  Yu-Ju Lin ,  Jiann-Hwa Liou

IPC分类号： G06F13/38 ,  H04L12/00 ,  H04L12/931 ,  H04L12/935 ,  H04L12/54

CPC分类号： H04L49/3027 ,  H04L49/201

摘要： A output queuing method for forwarding the packets sequentially in a switching network. Each port is related to a port output queue and a global output queue is shared by all port output queues. A FIFO (First In First Out) block is allocated into each port output queue and the global output queue. The FIFO block contains a number of FIFO nodes. Based on the type and destination ports of the received packet, the fields of the related FIFO node in the port output queue and the global output queue are set. The packets are sent out or skipped based on the related fields.

摘要翻译： 一种用于在交换网络中顺序转发分组的输出排队方法。 每个端口与端口输出队列相关，全局输出队列由所有端口输出队列共享。 FIFO（先进先出）块被分配到每个端口输出队列和全局输出队列中。 FIFO块包含多个FIFO节点。 根据接收到的数据包的类型和目标端口，设置端口输出队列中的相关FIFO节点和全局输出队列的字段。 根据相关字段发送或跳过数据包。

公开(公告)号：US20050094658A1

公开(公告)日：2005-05-05

申请号：US10960085

申请日：2004-10-08

申请人： Jen-Kai Chen ,  Chia-Hsin Chen

发明人： Jen-Kai Chen ,  Chia-Hsin Chen

IPC分类号： H04L12/56

CPC分类号： H04L49/201 ,  H04L49/351

摘要： A method for transmitting multi-cast packets is provided, including the steps of (a) receiving a plurality of multi-cast packets and generating corresponding port masks to indicate a plurality of ports from which the multi-cast packets are to be transmitted, (b) selecting, based on the yet-enqueued ports indicated in port masks, a port with highest priority, enqueuing the multi-cast packet into the port, and updating the port masks, (c) checking whether the port with highest priority being full-duplex, and (d) determining when to repeat steps (b) to (d) base on the result of the checking, until finishing the transmission specified by the port masks. In step (d), when the port with highest priority is full-duplex, wait until the port with highest priority starting transmitting the packet, then repeat steps (b) to (d). On the other hand, when the port with highest priority is half-duplex, wait until the port with highest priority transmitting a certain portion of the packet, for example 64 bytes or 128 bytes, then repeat steps (b) to (d). This will repeat until finishing all the transmission specified by the port masks. Preferably, the port with highest priority means the port of the highest priority group with the shortest output queue. When finishing all the transmission specified by the port masks, the switch will release all the related buffers of the multi-cast packets.

摘要翻译： 提供了一种用于发送多播分组的方法，包括以下步骤：（a）接收多个多播分组并生成相应的端口掩码，以指示要从中发送多播分组的多个端口（ b）根据端口掩码中指定的入口端口选择具有最高优先级的端口，将多播数据包排入端口，并更新端口掩码，（c）检查优先级最高的端口是否满 -duplex，以及（d）根据检查结果确定何时重复步骤（b）至（d），直到完成由端口掩码指定的传输。 在步骤（d）中，当优先级最高的端口为全双工时，请等待具有最高优先级的端口开始发送数据包，然后重复步骤（b）至（d）。 另一方面，当具有最高优先级的端口是半双工时，等待直到具有最高优先级的端口发送分组的某一部分，例如64字节或128字节，然后重复步骤（b）至（d）。 这将重复，直到完成由端口掩码指定的所有传输。 优选地，优先级最高的端口表示具有最短输出队列的最高优先级组的端口。 当完成端口掩码指定的所有传输时，交换机将释放多播数据包的所有相关缓冲区。

公开(公告)号：US06615336B1

公开(公告)日：2003-09-02

申请号：US09481713

申请日：2000-01-12

申请人： Jen-Kai Chen ,  Wei-Pin Chen ,  Jiann-Hwa Liou

发明人： Jen-Kai Chen ,  Wei-Pin Chen ,  Jiann-Hwa Liou

IPC分类号： G06F1202

CPC分类号： H04L12/413 ,  H04L45/745 ,  H04L49/351 ,  H04L49/90

摘要： A method for performing a MAC address lookup in a network switch of an Ethernet network is provided, using a memory structure including a number of multi-slot buckets. Each of the multi-slot buckets is mapped to a hash key derived from a MAC Address. Each of the multi-slot buckets has a number of forwarding table section, each storing a number of the forwarding table entries. If two forwarding table entries are stored in each forwarding table section of a 4-slot bucket ¼ memory space and ½ access time can be saved. Thus, the lookup for the forwarding table according to the MAC address in the Ethernet switch IC can be effectively performed.

摘要翻译： 提供了一种用于在以太网网络的网络交换机中执行MAC地址查找的方法，其使用包括多个多时隙桶的存储器结构。 每个多时隙桶被映射到从MAC地址导出的散列密钥。 每个多时隙桶具有多个转发表部分，每个转发表部分存储多个转发表条目。 如果两个转发表项存储在4槽桶¼存储空间的每个转发表部分中，并且可以保存½访问时间。 因此，可以有效地执行根据以太网交换机IC中的MAC地址的转发表的查找。

公开(公告)号：US20110059628A1

公开(公告)日：2011-03-10

申请号：US12553991

申请日：2009-09-04

申请人： Jen-Kai Chen

发明人： Jen-Kai Chen

IPC分类号： H01R12/00

CPC分类号： H01R31/065 ,  H01R27/02

摘要： The present invention constructs a SD Flash card by plugging two micro-SD cards into a new apparatus that has the same form factor as a SD Flash card. In this new apparatus, there is a controller to bridge the two micro-SD cards of any SD interface speed type (DS, HS, UHS50 or UHS104) to UHS104. The controller performs striping access function to achieve almost double performance in sequential read/write throughput if it is not limited by the target SD interface speed.

摘要翻译： 本发明通过将两个micro-SD卡插入具有与SD闪存卡相同形状的新设备来构造SD闪存卡。 在这种新设备中，有一个控制器将任何SD接口速度类型（DS，HS，UHS50或UHS104）的两个micro-SD卡桥接到UHS104。 如果不受目标SD接口速度的限制，控制器将执行条带访问功能，以实现连续读/写吞吐量的几乎两倍的性能。

公开(公告)号：US07382728B2

公开(公告)日：2008-06-03

申请号：US10157238

申请日：2002-05-30

申请人： Jen-Kai Chen ,  Hsiang-Yi Huang

发明人： Jen-Kai Chen ,  Hsiang-Yi Huang

IPC分类号： H04L12/56

CPC分类号： H04L47/10 ,  H04L47/2433 ,  H04L47/50 ,  H04L47/6215

摘要： A network switching apparatus and method for congestion control. Each one of the connection ports of the switching apparatus includes a low priority queue and a high priority queue. When a data packet enters a switching apparatus, the switching apparatus according to the type of the data packet enqueues the data packet to the low priority queue or the high priority queue. When congestion occurs at the switching apparatus, the low priority queue and the high priority queue can respectively perform the different ways of the congestion control, according to the input congestion mode. Also, when the switching apparatus receives a pause frame, response flow control can be performed, according to the output congestion mode. Since the different ways of the congestion control are performed according to the different types of the data packet, the congestion control can be optimized.

摘要翻译： 一种用于拥塞控制的网络交换设备和方法。 交换设备的每个连接端口包括低优先级队列和高优先级队列。 当数据分组进入交换设备时，根据数据分组的类型的交换设备将数据分组引入低优先级队列或高优先级队列。 当交换装置发生拥塞时，根据输入拥塞模式，低优先级队列和高优先级队列可以分别执行拥塞控制的不同方式。 此外，当切换装置接收到暂停帧时，可以根据输出拥塞模式执行响应流控制。 由于根据不同类型的数据包执行拥塞控制的不同方式，所以可以优化拥塞控制。

公开(公告)号：US06954425B2

公开(公告)日：2005-10-11

申请号：US09855713

申请日：2001-05-16

申请人： Jen-Kai Chen ,  Chien-Shu Tseng ,  Jiann-Hwa Liou

发明人： Jen-Kai Chen ,  Chien-Shu Tseng ,  Jiann-Hwa Liou

IPC分类号： G06F1/10 ,  G06F5/06 ,  H04J3/06 ,  H04J1/16

CPC分类号： G06F5/06 ,  G06F1/10 ,  H04J3/0685

摘要： A method for compensating for clock signal difference between a switch and peripheral device, including a receiving process and a transmitting process. For the receiving process, after an N-th packet is received, a first counter is triggered and begins to count. When an (N+1)-th packet is inputted, the counter stops counting and then an inter-packet gap IPG(N, N+1) between the N-th packet and the (N+1)-th packet is recorded into the N+1 queue link node QLN(N+1) according a counting value by the first counter; otherwise, the first counter keeps counting. For the transmitting process, after an inter-packet gap IPG(M−1, M) is obtained, and the M-th packet is transmitted, and then the second counter is triggered to begin to count. When a counted value by the second counter is equal to the clock cycle value corresponding to the inter-packet gap IPG(M−1, M), the second counter stops counting. When an inter-packet gap IPG(M, M+1) is obtained, and the (M+1)-th packet is then transmitted; otherwise, the second counter keeps counting.

摘要翻译： 一种用于补偿交换机和外围设备之间的时钟信号差异的方法，包括接收过程和发送过程。 对于接收处理，在接收到第N个分组之后，触发第一个计数器并开始计数。 当输入第（N + 1）个分组时，计数器停止计数，然后记录第N个分组和第（N + 1）个分组之间的分组间间隔IPG（N，N + 1） 根据第一计数器的计数值，将N + 1队列链接节点QLN（N + 1） 否则，第一个计数器继续计数。 对于发送处理，在获得分组间间隔IPG（M-1，M）之后，发送第M个分组，然后触发第二计数器开始计数。 当第二计数器的计数值等于对应于分组间间隔IPG（M-1，M）的时钟周期值时，第二计数器停止计数。 当获得分组间间隔IPG（M，M + 1）时，然后发送第（M + 1）个分组; 否则，第二个计数器继续计数。

公开(公告)号：US06697923B2

公开(公告)日：2004-02-24

申请号：US09987499

申请日：2001-11-15

申请人： Jen-Kai Chen ,  ChaoCheng Cheng ,  Yeong-Chian Hu

发明人： Jen-Kai Chen ,  ChaoCheng Cheng ,  Yeong-Chian Hu

IPC分类号： G06F1200

CPC分类号： G06F5/10

摘要： A method for buffer management and a controller of the same are disclosed. In the buffer management method, a first control mode is performed, and a plurality of bits are used to control a bit mask region of a memory. A second control mode is performed, and a plurality of unused addresses in a link region of the memory is cached. A third control mode is performed, to control a plurality of second unused addresses in the link region by a linked list. The controller comprises a plurality of bits for controlling a bit mask region; a plurality of address cache units for caching a plurality of first unused address of a linked list in a link region; and a pointer for always pointing to a head of the linked list in the link region, wherein the linked list links a plurality of second unused addresses.

摘要翻译： 公开了一种用于缓冲器管理的方法及其控制器。 在缓冲器管理方法中，执行第一控制模式，并且使用多个位来控制存储器的位掩码区域。 执行第二控制模式，并且存储器的链接区域中的多个未使用的地址被缓存。 执行第三控制模式，以通过链表控制链路区域中的多个第二未使用地址。 控制器包括用于控制位掩码区域的多个位; 多个地址缓存单元，用于在链接区域中缓存链表的多个第一未使用地址; 以及用于总是指向所述链接区域中的所述链接列表的头部的指针，其中所述链接列表链接多个第二未使用的地址。

公开(公告)号：US07310311B2

公开(公告)日：2007-12-18

申请号：US10384697

申请日：2003-03-11

申请人： Jen-Kai Chen ,  Hsiao-Lung Wu ,  Hsiang-Yi Huang

发明人： Jen-Kai Chen ,  Hsiao-Lung Wu ,  Hsiang-Yi Huang

IPC分类号： G01R31/08 ,  H04J1/16 ,  H04L1/00

CPC分类号： H04L49/351 ,  H04L49/50

摘要： An Ethernet switch with rate control and associated method is provided. Each port in the switch has individual settings of egress/ingress) rate control, which are stored in a register and configured based on required rates. The switch uses data volume that a port can output/input within each unit time to control egress/ingress rate of the port. Further, the egress rate can be precisely controlled by using uniform random numbers provided by an random number generator of the switch, and the ingress rate can be advantageously controlled by combining a proper kind of congestion control, which is performed according to the capability of a device connected to the port, such as full-duplex or half-duplex, and flow control.

摘要翻译： 提供了一种具有速率控制和相关方法的以太网交换机。 交换机中的每个端口都有单独的出口/入口设置）速率控制，它们存储在寄存器中，并根据所需的速率进行配置。 交换机使用端口可以在每个单位时间内输出/输入的数据量，以控制端口的出口/入口速率。 此外，可以通过使用由交换机的随机数发生器提供的均匀随机数来精确地控制出口速率，并且可以有利地通过组合适当类型的拥塞控制来控制入口速率，所述拥塞控制是根据 连接到端口的设备，如全双工或半双工，以及流量控制。