Abstract: A data-packet processing method is used in a network system. The network system includes a buffer for optionally storing a data packet to be transferred, and the method includes steps of: detecting a remaining space of the buffer; prohibiting a first type of data packet and a second type of data packet from being stored into the buffer when the remaining space of the buffer is equal to or less than a first threshold; allowing the first type of data packet and the second type of data packet to be stored into the buffer when the remaining space of the buffer is greater than a second threshold; and allowing the first type of data packet to be stored into the buffer while prohibiting the second type of data packet from being stored into the buffer when the remaining space of the buffer is greater than the first threshold but equal to or less than the second threshold.

Abstract: Systems and methods for broadcast storm control. A system for broadcast storm control comprises a plurality of ports and a processing module. Each port comprises a counter counting the number of broadcast packets received in a time unit, and a corresponding drop flag set based on the counter value and an ingress rate thereof. The processing module receives a packet from a specific port within the ports, and drops or forwards the packet to at least one of the ports according to the drop flag corresponding to the specific port.

Abstract: Systems and methods for packet forward control. The system comprises a plurality of ports and a processing module configured to receive a packet. The processing module comprises a resource management module and a forward control module. The resource management module detects whether congestion occurs on at least one specific port among the ports. The forward control module determines whether the specific port comprises a predetermined port within the ports, and if so, forwards the packet to at least one of the ports excepting the predetermined port.

Abstract: The invention provides a method for handling the bandwidth allocation of a networking device. The networking device forwards a plurality of packets. A host attribute table is first searched according to a host attribute of the packets to obtain a host attribute table index of the packets. An application attribute table is then searched according to an application attribute of the packets to obtain an application attribute table index of the packets. A bandwidth profile table is then searched according to both the host attribute table index and the application attribute table index of the packets to obtain a bandwidth profile of the packets. Finally, the packets are limited to a specific bandwidth according to the bandwidth profile of the packets.

Abstract: The invention provides a service VLAN tag format, according to that a service VLAN tag header is appended by a provider network to a plurality of packets forwarded by the provider network. The service VLAN tag format comprises a 2-byte tag protocol identifier (TPID) field and a 2-byte tag control information (TCI) field. The TPID field comprises a (16-N)-bit shortened TPID field indicating that a service VLAN tag header is following, and an N bit extra VLAN identifier field. The TCI field comprises a 3-bit user priority field indicating a packet priority, a 1-bit canonical format indicator (CFI) field, and a 12-bit original VLAN identifier field. The extra and original VLAN identifier fields are combined to indicate a service VLAN identifier that is (12+N) bits in length.

Abstract: A packet switching device simultaneously supports a tagged mode and an untagged mode. A plurality of ports of the packet switching device perform packet delivery and reception, and a port map defines VIDs and operating modes of the ports. Each port operates in tagged or untagged mode, and is associated with at least one VID. The range of VIDs in untagged mode is bounded in a first range, the range of VIDs in tagged mode is bounded in a second range, and the first and second ranges are not overlapped. The plurality of ports comprises a first port operating in untagged mode, associated with a first VID within the first range. The first port receives a first packet, and the packet switching device searches the port map to locate other ports having the first VID and forwards the first packet thereto.

Abstract: A method for MAC address management is provided. A MAC address is provided, and capacities of a first column and a second column corresponding to a first index in a MAC table are checked. If the first column and the second column are filled, capacities of a third column and a fourth column corresponding to a second index in the MAC table are checked. If one of the third and fourth columns is empty, the MAC address is written thereto. The second index is successive to the first index. The MAC table further comprises a plurality of time stamp columns, recording idle time of each MAC address.

Abstract: A control system for packet transmission. A forward control unit receives a packet providing a port count through an input port and implements a lookup operation. A queue control unit determines whether the port count is greater than a first predetermined value, and, if not, outputs the packet through a first port, and, if so, sends a first message. A multicast forward control unit receives the first message, adds one to a first packet count, determines whether the first packet count is greater than a threshold value, and, if so, sends a stop forward message to the forward control unit but does not send a grant message. The forward control unit receives the stop forward message and does not transmit subsequent packets to the first port but to a second port directly.

Abstract: A packet transmission method. A switch, a server, and at least one host are provided. The switch comprises at least one first port and one second port and the host connects to the switch via the first port and the server connects to the switch via the second port. The server transmits a frame with a VID value to the host via the second port. It is determined whether the first port and the second port are the same group members. If so, the frame is forwarded to the first port. It is then determined whether the VID value of the frame is acceptable by the first port. If unacceptable, the VID value of the frame is swapped with an acceptable VID value, and the host receiving the frame with the swapped VID value via the first port.

Abstract: A method for implementing varying service quality grades in a network switch is provided, comprises the steps of: classifying a plurality of users of the network switch into a plurality of service quality grades according to a contributing factor of the plurality of the users; connecting each of the plurality of users to the network switch via one of a plurality of ports of the network switch according to its service quality grade; determining an original priority of a packet in the network switch; and determining an adjusted priority of the packet according to both the original priority of the packet and a priority adjustment table of an ingress port of the packet. Each of the plurality of ports has a corresponding priority adjustment table, and each of the priority adjustment tables includes a mapping relationship between the original priority and the adjusted priority.

Abstract: A method for implementing packet en-queuing and de-queuing processes in a network switch is provided. The method comprises the following steps. First, an en-queuing process and a de-queuing process are divided into a plurality of en-queuing and de-queuing stages. The en-queuing process of a plurality of en-queued packets is then processed with each of the plurality of en-queued packets processed in one of the plurality of en-queuing stages simultaneously, and every one of the plurality of en-queued packets passes through all of the plurality of en-queuing stages sequentially to complete the en-queuing process. The de-queuing process of a plurality of de-queued packets is then processed with each of the plurality of de-queued packets processed in one of the plurality of de-queuing stages simultaneously, and every one of the plurality of de-queued packets passes through all of the plurality of de-queuing stages sequentially to complete the de-queuing process.

Abstract: A method for processing packets of a VLAN in a network switch is provided. The VLAN comprises a plurality of physical LANs and are divided into a plurality of subnets. The method comprises the steps of: receiving a packet through an ingress port of the network switch, wherein the packet is tagged with a VID of the VLAN; finding a VLAN member according to the VID, wherein the VLAN member represents all ports of the network switch dedicated to serving packets of the VLAN; finding a forwarding scope according to the ingress port, wherein the ingress port is one of the ports dedicated to serving packets of a first subnet of the VLAN and the forwarding scope represents all ports of the network switch dedicated to serving packets of the first subnet; and determining an egress port of the packet according to both the VLAN member and the forwarding scope.

Abstract: The present invention relates to a structure and a method of data update in a cache memory inside a local processor, which uses the feature of cache control. A buffer block of a header buffer is mapped to a memory space at several different address sectors addressed by the local processor. Whenever the local processor attempts to access the internal cache memory, cache missing will occur so that a local processor will be forced to alternatively request new data from buffer blocks of a header buffer in a HCA. Consequently, the whole block is loaded into cache memory. This does not only boost cache update performance but also accelerates packet access.

Abstract: A symmetric dual-slot data hash method and a switching apparatus using the same having a first block and a second block of equal size for storing MAC addresses. The size of the dual-slot address table is p×2m with m is an integer, and p is 2 or a prime. The hash address comprises a higher address portion and a lower address portion. A CRC hash operation is performed to the MAC address by using the m-bit CRC function to obtain a hash value. Next, the highest bit of the hash value is removed and the remained bits are used as the lower address portion of the hash address. A dividend is set according to the m-th bit of the MAC address. After the dividend is divided by a divisor p to obtain a remainder, the remainder is set as a higher address portion of the hash address.

Abstract: A drive mechanism for use in a man-driven vehicle having a front fork to which a wheel is rotatably connected includes a rack, a power drive and a quick-release lock. The rack is affixed to the front fork of the man-driven vehicle. The power drive has a transmission shaft pressable on a periphery of the wheel of the front fork, and a motor coupled to the transmission shaft for driving the transmission shaft to rotate. The quick-release lock is coupled between said power drive and said rack for locking said power drive and said rack together, thereby enabling the user to select the man power driving mode or motor driving mode.

Abstract: A relay control method and circuit is designed for use on a computer network system, such as Ethernet, which can perform a load-balancing operation based on a port group configuration in such a manner as to allow an increased network data communication efficiency through an Ethernet Switch. In the invention, a CRC (Cyclic Redundancy Check) operation is performed on the binary address information extracted from each received frame to thereby obtain a CRC modulo. Based on the CRC modulo, the target port of the received data frame is found from a routing table. Then the received frame is forwarded according to the index address, the target port, and the port group configuration. The port group configuration can be adjusted if an over-loading is occurred in the ports belonging to a port group.

Abstract: A drive mechanism for use in a man-driven vehicle having a front fork to which a wheel is rotatably connected includes a rack, a power drive and a quick-release lock. The rack is affixed to the front fork of the man-driven vehicle. The power drive has a transmission shaft pressable on a periphery of the wheel of the front fork, and a motor coupled to the transmission shaft for driving the transmission shaft to rotate. The quick-release lock is coupled between said power drive and said rack for locking said power drive and said rack together, thereby enabling the user to select the man power driving mode or motor driving mode.

Abstract: A relay control circuit that uses a hashing algorithm for increasing the hit rate in a table lookup operation. Destination medium access control (DMAC) address is extracted from a network frame. The DMAC address is scrambled to put the frequently changing bytes in a position closer to each other. The scrambled address is sent to a cyclic redundancy checking circuit, where a cyclic redundancy checking operation with hashing function capability is carried out to produce a remainder. The remainder is sent to a memory unit for a table lookup operation. Hence, the relay control circuit is capable of finding a destination port that corresponds to a given DMAC address in a very short time, thereby boosting network efficiency.

Abstract: The present invention relates to a structure and a method of data update in a cache memory inside a local processor, which uses the feature of cache control. A buffer block of a header buffer is mapped to a memory space at several different address sectors addressed by the local processor. Whenever the local processor attempts to access the internal cache memory, cache missing will occur so that a local processor will be forced to alternatively request new data from buffer blocks of a header buffer in a HCA. Consequently, the whole block is loaded into cache memory. This does not only boost cache update performance but also accelerates packet access.

Abstract: A message transmitting queue delivers messages between a source controller and a destination controller. According to the message transmitting request of the source controller, sequentially distribute the free message row of the message transmitting queue, and set the message row to the distributed state. After the source controller writes the message of the message row, set the message row to the written state. At this moment, when the message row is in the position that is read sequentially by the destination controller, a read request is issued, so that the destination controller reads the message according to the read request when the reading completes, clears the distributed signal and the written signal, so that the message row goes back to the free state. When the message transmitting queue has no free message row, a no free message row signal to inform the source controller is issued.