Migrating enterprise IT services to the cloud is becoming a trend. However, configuring virtual networks in the cloud is a costly and error-prone task. In this paper, we model the problem of VPC intent compilation and prove that the problem is NP-hard. We design a heuristic method to automatically translate the intents into VPC configurations. To facilitate the heuristic, we propose an algorithm for finding small cuts (AFSC) based on the Louvain algorithm, which is used to separate subnets across VPCs. The generated configurations find an equilibrium between maximizing the network performance and minimizing the resource consumption. Experimental results show that our heuristic method reaches 100% correctness, compiles large intent sets from real-world networks with hundreds of subnets and thousands of intents in under a minute, and improves the effectiveness by ∼2.8× in terms of resource consumption relative to network performance.

Network programming languages (NPLs) empower operators to program network data planes (NDPs) with unprecedented efficiency. Currently, various NPLs and NDPs coexist and no one can prevail over others in the short future. Such diversity is raising many problems including: (1) programs written with different NPLs can hardly interoperate in the same network, (2) most NPLs are bound to specific NDPs, hindering their independent evolution, and (3) compilation techniques cannot be readily reused, resulting in much wasteful work. These problems are mostly owing to the lack of modularity in the compilers, where the missing part is an intermediate representation (IR) for NPLs. To this end, we proposeNetwork Transaction Automaton (NTA), a highly-expressive and language-independent IR, and show it can express semantics of 7 mainstream NPLs. Then, we designCODER, a modular compiler based on NTA, which currently supports 2 NPLs and 3 NDPs. Experiments with real and synthetic programs show CODER can correctly compile those programs for real networks within moderate time.