Abstract: A connection management server apparatus is arranged for the plurality of data link activated side apparatuses to systematically manage information of logical link resources to be used to connect the Internet. When a data link activating side apparatus newly calls a data link activated side apparatus, the connection management server apparatus provides information that represents a free logical link resource and a data link activated side apparatus having the logical link resource to the data link activated side apparatus that is newly called.

Abstract: In a remote access server (RAS), the number of logical link resources, which are circuit resources between the RAS and an ISP server, is set greater than the number of physical link resources, which are circuit resources between the RAS and each terminal device. In the RAS, when a particular circuit undergoes a transition to the dormant state, only the physical link resource is released to allow use by other terminal devices while the logical link resource is maintained unchanged in a connected state, whereby the loss probability is reduced without increasing the number of physical link resources.

Abstract: A monitor (102) receives a returned response message, and stores the state of the link directed to the NE that returned the message as a successful transmission, and then repeats the same operation at arbitrary intervals. When the monitor can not receive the response message after placing a certain interval, it stores the link as an unsuccessful transmission. At this point, if all the links for a particular NE are in the unsuccessful state, the monitor informs an NMS (151) of a fault in that NE. If the NE truly has the NE fault, other NEs that also have the links to that fault NE inform the NMS of the fault. Thereby the NMS itself does not need to do the monitoring. The NMS only collates fault information with topology information stored in a topology information storage (153) and can judge that a particular NE is out of communication and control.

Abstract: In a remote access server (RAS), the number of logical link resources, which are circuit resources between the RAS and an ISP server, is set greater than the number of physical link resources, which are circuit resources between the RAS and each terminal device. In the RAS, when a particular circuit undergoes a transition to the dormant state, only the physical link resource is released to allow use by other terminal devices while the logical link resource is maintained unchanged in a connected state, whereby the loss probability is reduced without increasing the number of physical link resources.

Abstract: A communication system includes a plurality of data link activating side apparatuses, a plurality of data link activated side apparatuses, and a connection management server apparatus. The data link activating side apparatuses use the Internet. The data link activated side apparatuses are connected to the data link activating side apparatuses through a circuit switched network to relay data to be transmitted/received between the data link activating side apparatuses and the Internet. The connection management server apparatus is arranged for the plurality of data link activated side apparatuses to systematically manage information of logical link resources to be used to connect the Internet.

Abstract: A data communication system includes a server, data link started side apparatus, data link starting side apparatus, and originating server. The server provides a service through a network. The data link started side apparatus performs line control between itself and the line connection destination, and performs LAN control between itself and the server through the network, thereby relaying packets between the line connection destination and the server. The data link starting side apparatus is connected to the data link started side apparatus, disconnects the line between itself and the data link started side apparatus on the basis of an MITF dial-up dormant protocol when there is no communication between itself and the server for a predetermined period of time, and shifts to a dormant state.

Abstract: A monitor (102) receives a returned response message, and stores the state of the link directed to the NE that returned the message as a successful transmission, and then repeats the same operation at arbitrary intervals. When the monitor can not receive the response message after placing a certain interval, it stores the link as an unsuccessful transmission. At this point, if all the links for a particular NE are in the unsuccessful state, the monitor informs an NMS (151) of a fault in that NE. If the NE truly has the NE fault, other NEs that also have the links to that fault NE inform the NMS of the fault. Thereby the NMS itself does not need to do the monitoring. The NMS only collates fault information with topology information stored in a topology information storage (153) and can judge that a particular NE is out of communication and control.