On M-concurrency path computation and its application in dynamic service Multi-Layer Networks

Dynamic path computation and optimization is very challenging for emerging Multi-Layer Networks (MLN), typically Packet Optical Transport Networks (P-OTN). In this work, we propose to address a category of dynamic optimization problems in MLNs by using an M-concurrency path computation approach. By leveraging our previous work in single path computation solutions for MLNs, we design an efficient M-concurrency path computation heuristic to optimize dynamic service provisioning. Based on this base solution, we develop optimization schemes to address two realistic application problems: (1) dynamic optimization for service reprovisioning and (2) dynamic optimization for mixed scheduling and immediate requests. Experimental results are presented for performance evaluation and comparison of the investigated optimization solutions and schemes.