In future 5G infrastructures, network services will be deployed through sets of Virtual Network Functions (VNFs) leveraging the advantages of both Software Defined Networking (SDN) and Network Function Virtualization (NFV). A network service is composed of an ordered sequence of VNFs, i.e., VNF Forwarding Graph (VNFFG), deployed across distributed data centers (DCs). Herein, we present a Cloud/Network Orchestrator which dynamically processes and accommodates VNFFG requests over a pool of DCs interconnected by a multi-layer (packet/flexi-grid optical) transport network infrastructure. We propose two different cloud and network resource allocation algorithms aiming at: (i) minimizing the distance between the selected DCs, and (ii) minimizing the load (i.e., consumed cloud resources) of the chosen DCs. Both algorithms run on a Cloud/Network Orchestrator and are experimentally validated and benchmarked on the CTTC ADRENALINE testbed.