Compute First Networking: Distributed Computing meets ICN

Abstract

Modern distributed computing frameworks and domain-specific languages provide a convenient and robust way to structure large distributed applications and deploy them on either data center or edge computing environments. The current systems suffer however from the need for a complex underlay of services to allow them to run effectively on existing Internet protocols. These services include centralized schedulers, DNS-based name translation, stateful load balancers, and heavy-weight transport protocols. In contrast, ICN-oriented remote invocation methodologies provide an attractive match for current distributed programming languages by supporting both functional programming and stateful objects such as Actors. In this paper we design a computation graph representation for distributed programs, realize it using Conflict-free Replicated Data Types (CRDTs) as the underlying data structures, and employ RICE (Remote Method Invocation for ICN) as the execution environment. We show using NDNSim simulations that it provides attractive benefits in simplicity, performance, and failure resilience.