Design And Implementation Of Login And Management Module For N Nodes Of NPV Switch System

Because of the number of Domain ID(each switch is assigned a Domain ID, and the number of valid Domain ID is 239), the scale of Fabric network is very small which composed by FC Switch, Each SAN (or VSAN) supports a maximum of 239 domain IDs, so the SAN has a limit of 239 switches. In a SAN topology with a large number of edge switches, the SAN may need to grow beyond this limit. NPV mode alleviates the domain ID limit by sharing the domain ID of the core switch among multiple edge switches.In NPV mode, the edge switch relays all traffic from server-side ports to the core switch. The edge switch shares the domain ID of the core switch. An NP uplink is a connection from an NP port on the edge switch to an F port on the core switch. Server interfaces are F ports on the edge switch that connect to the servers. The core switch provides F port functionality and all the Fibre Channel switching capabilities. The edge switch appears as a Fibre Channel host switch to the core switch and as a regular Fibre Channel switch to its connected devices.N nodes login by sending FLOGIN or FDISC packets. When sending FLOGIN packets, the two points of link will negotiate the attribute on the link, such as flow control type, the length of flow control parameters,size of the temporary buffer, BBCredit, whether to support NPIV, Whether to negotiate VSAN. After the negotiate, the N nodes will be assigned a FCID to complete login. Both FLOGIN and FDISC will be used to request FCID. But, when sending FLOGIN to request FCID, all of the login information will be logout on which interface the FLOGIN be receive, and then the FLOGIN will be process. When sending FDISC to request FCID, Both the sending and rcv must to support NPIV, and the interface which receive FDISC should have more than one login information.In order to improve the stability and scalability of the switch, the switch has a capability of distributed processing. And this cause the switch will be able to login and logout with distributed environment, And some information need to be passed between the plates such as the flag of debugging, the flag of NPV enable, mapping, login DB, FDISC proxy and so on, so need the mechanism to process this.