NeTS: SMALL: Develop Core Techniques and Applications of Learned and Disaggregated In-Network Lookups

Table lookups serve as fundamental functions and design blocks of numerous computer network protocols and algorithms. With the emergence of new network application scenarios such as cloud and edge computing paradigms, the Internet of Things (IoT), smart communities, self-driving vehicles, and distributed machine learning, modern networks have been using in-network lookups beyond classic Internet Protocol (IP) or Medium Access Control (MAC) address forwarding. The project plans to investigate a new paradigm using machine learning models to replace traditional hash functions in lookup engines and disaggregating lookup functions for heterogeneous devices. The project aims to develop LEarned and Disaggregated In-network Lookup Engines (LEDILE), a next-generation network framework that provides cost efficiency, high performance, effectiveness in handling failures, scalability to large networks, and compatibility with emerging network features. This project seeks to fundamentally change the design and deployment of classic in-network lookup engines by replacing a functional stage with a learned model and disaggregating lookup functions onto heterogeneous devices. The proposed research will include the following: 1) Develop new in-network lookups with perfect hashing and learned models; 2) Disaggregate in-network lookups in two dimensions: stages and shards; 3) Develop the LEDILE framework and its applications; 4) Evaluate the proposed algorithms, protocols, and software framework on multiple platforms. If successful, the research outcomes of this project will be transformative as they will provide critical networking functions and services in emerging networks including cloud, IoT, edge computing, and their applications. The PI plans to integrate the research being conducted under this project into the undergraduate and graduate curriculum. The algorithms, protocols, software, and experimental tools developed in this project will be made available to the public to enable other researchers to work in this area. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.