Abstract-Cloud computing integrates data, applications, users and servers on a vast scale and enables a global optimization of computing resources. However, due to security threats from both outside and inside the cloud, security remains as a significant challenge and obstacle in the wide adoptions of cloud computing paradigms. To enhance the security of networks, applications and data in the cloud, this position paper proposes to develop a profiling-as-a-service architecture to characterize, understand and profile network traffic at multiple layers in the multi-tenant cloud computing environment: network routers, hypervisors, virtual instances and applications. The proposed architecture will not only provide an in-depth understanding on traffic patterns of cloud tenants, but also enhance the security of cloud computing by collaboratively detecting and filtering unwanted traffic towards cloud instances.