Management architecture against hardware failures in an optical packet and circuit integrated network

We design and develop architecture against hardware failures arisen on an optical packet and circuit integrated (OPCI) network. The control plane regularly collects hardware failures´ information from the data plane, automatically notifies the failures to the management plane (M-plane), and recovers the status of path usage. For the M-plane, we develop and install a management system based on a hierarchical management model constructed as a relational database to visually manage various OPCI network elements and their correlative relationships over multiple layers. The M-plane contributes to faster discovery of resources affected by hardware failures than the existing OPCI network without any management systems. In verification experiments, we show that hardware failures and resources affected by the failures are successfully displayed on the management window. Besides, the empirically-estimated time to detect a hardware failure and specify affected resources indicates indispensability of the M-plane in that the network administrator can work on recovery operations much faster and earlier and enhance availability of the OPCI network.