Abstract: The federated learning system includes a moderator and client devices. Each client device performs a method for verifying model update as follows: receiving a hash function and a general model; training a client model according to the general model and raw data; calculating a difference as an update parameter between the general model and the client model, sending the update parameter to the moderator; inputting the update parameter to the hash function to generate a hash value; sending the hash value to other client devices, and receiving other hash values; summing all the hash values to generate a trust value; receiving an aggregation parameter calculated according to the update parameters; inputting the aggregation parameter to the hash function to generate a to-be-verified value; and updating the client model according to the aggregation parameter when the to-be-verified value equals the trust value.

Abstract: A display apparatus includes a multiplexer circuit, a driving unit, a first control line and a second control line. The multiplexer circuit includes a plurality of switch units. The first control line is electrically connected with the switch units and the driving unit. The second control line is electrically connected with the switch units and the driving unit. A maximum time constant from the driving unit to the switch units is less than R*C/4, wherein R represents the equivalent resistance of the portion of the first control line between the two switch units which are the farthest from each other, and C represents the equivalent capacitance of the portion of the first control line between the two switch units which are the farthest from each other.

Abstract: A display apparatus includes a multiplexer circuit, a driving unit, a first control line and a second control line. The multiplexer circuit includes a plurality of switch units. The first control line is electrically connected with the switch units and the driving unit. The second control line is electrically connected with the switch units and the driving unit. A maximum time constant from the driving unit to the switch units is less than R*C/4, wherein R represents the equivalent resistance of the portion of the first control line between the two switch units which are the farthest from each other, and C represents the equivalent capacitance of the portion of the first control line between the two switch units which are the farthest from each other.

Abstract: A load balancing routing method for networks is disclosed. The routing method includes following steps. A network topology graph and a plurality of expected bandwidth demands corresponding to a plurality of source-destination pairs are received by a network server. A plurality of link criticalities of a plurality of links established according to the source-destination pairs is calculated by the network server according to the network topology graph and the expected bandwidth demands. A plurality of expected loads of the links is calculated by the network server according to the link criticalities. A plurality of cost values is calculated according to a plurality of residual capacities of the links and the corresponding expected loads. A better transmission path corresponding to each of the source-destination pairs is selected by the network server according to the weighted sum of the cost values corresponding to the links in the source-destination pair.

Abstract: A data center network system and a packet forwarding method are provided. The data center network includes a management server and a plurality of machines containing physical machines and virtual machines. The management server configures a logical media access control (MAC) address for each of the machines, wherein most significant bytes of each of the logical MAC addresses are set as 0. When a data packet is about to be sent from a physical machine, the physical machine executes an encapsulation procedure on the data packet for forwarding the data packet to an intermediate node between a transmitter and a receiver of the data packet, and the intermediate node executes a decapsulation procedure on the data packet for forwarding the data packet to the true receiver. Accordingly, the number of virtual machines exposed to the forwarding table of Ethernet switches can be effectively reduced.

Abstract: A network system and a method of address resolution are provided. The network system includes a network, a plurality of virtual machines, a routing module and a path directory module. Each virtual machine includes an internet protocol (IP) address and N media access control (MAC) addresses, so as to connect the network through N transmission routes. The routing module detects and calculates states of the transmission routes. The path directory module receives and decodes an address resolution protocol (ARP) request presented by at least one source virtual machine to reply path information, which includes N MAC addresses corresponding to an IP address of a destination virtual machine and states of the N transmission routes. Thus, the virtual machines present the ARP request without broadcast, so that the problem of network congestion is solved.

Abstract: A network system and a method of address resolution are provided. The network system includes a network, a plurality of virtual machines, a routing module and a path directory module. Each virtual machine includes an internet protocol (IP) address and N media access control (MAC) addresses, so as to connect the network through N transmission routes. The routing module detects and calculates states of the transmission routes. The path directory module receives and decodes an address resolution protocol (ARP) request presented by at least one source virtual machine to reply path information, which includes N MAC addresses corresponding to an IP address of a destination virtual machine and states of the N transmission routes. Thus, the virtual machines present the ARP request without broadcast, so that the problem of network congestion is solved.

Abstract: A load balancing routing method for networks is disclosed. The routing method includes following steps. A network topology graph and a plurality of expected bandwidth demands corresponding to a plurality of source-destination pairs are received by a network server. A plurality of link criticalities of a plurality of links established according to the source-destination pairs is calculated by the network server according to the network topology graph and the expected bandwidth demands. A plurality of expected loads of the links is calculated by the network server according to the link criticalities. A plurality of cost values is calculated according to a plurality of residual capacities of the links and the corresponding expected loads. A better transmission path corresponding to each of the source-destination pairs is selected by the network server according to the weighted sum of the cost values corresponding to the links in the source-destination pair.

Abstract: A data center network system and a packet forwarding method are provided. The data center network includes a management server and a plurality of machines containing physical machines and virtual machines. The management server configures a logical media access control (MAC) address for each of the machines, wherein most significant bytes of each of the logical MAC addresses are set as 0. When a data packet is about to be sent from a physical machine, the physical machine executes an encapsulation procedure on the data packet for forwarding the data packet to an intermediate node between a transmitter and a receiver of the data packet, and the intermediate node executes a decapsulation procedure on the data packet for forwarding the data packet to the true receiver. Accordingly, the number of virtual machines exposed to the forwarding table of Ethernet switches can be effectively reduced.