TinyNF

This repository contains the "TinyNF" driver codebase.

It was originally associated with the paper "A Simpler and Faster NIC Driver Model for Network Functions" (OSDI 2020). You can still find that version of the code, and the experimental scripts to reproduce that paper, in the osdi20 branch.

It was extended for the paper "Safe low-level code without overhead is practical" (ICSE 2023). The repo contains more programming languages, a second driver model, and scripts to reproduce the new paper's experiments. It is permanently archived at https://doi.org/10.5281/zenodo.7649472

Code

The code of the drivers is in the ada, c, csharp, and rust folders. We refer to "agents" in the code for the restricted TinyNF model and "queues" for the flexible DPDK model.

You can make each driver to compile it, and make format it to auto-format the code.

To run a driver with a minimal app, either manually run the binary produced by compilation, or make -f Makefile.benchmarking, but see the benchmarking/ folder for required environment variables.

The following environment variables are supported by each driver's Makefile:

• TN_MODE: The kind of driver:
  • restricted (default): The "restricted" model, which is the original TinyNF one
  • const (ada, c, and rust only): The restricted model with a constant number of devices, instead of detecting them at run-time
  • flexible: The "flexible" model, using queues similar to DPDK
  • noextensions (csharp only): The flexible model using only safe C#, without the language extensions
• TN_CSHARP_AOT (csharp only): Use ahead-of-time compilation for C# rather than the default just-in-time (any value given to the variable = AOT, undefined = JIT)

Note that despite needing extensions, the Rust driver does not support TN_MODE=safe because, due to Rust's ownership model, unsafe code must be used in the hot loop for volatile reads and writes, whereas C# allows these reads and writes in safe code.

All languages force inlining of the driver methods into a forced-not-inlined "run" method to ensure a fair comparison and to make extracting the hot loop code easy.

All languages use a 64-bit integer to represent packet lengths, even though the card only supports 16 bits, because not doing so often causes the card to lock up in non-C implementations. (Yes, this sounds extremely weird, but that's what empirical evidence says...)

Dependencies

To compile each language, you will need make, as well as a compiler:

• ada: gnat, though any other compiler might work
• c: gcc or clang, though any other C11 compiler should work
• csharp: dotnet, version 7 or above
• rust: rustc, a version that supports Rust 2021, and the cargo build system

If you don't want to install those on your machine, we provide a Dockerfile, just run docker build -t tinynf . ; docker run -it tinynf (you might need sudo for Docker). This file is also useful if you want to know how to install the dependencies on any Ubuntu machine. (The Dockerfile does not include clang-format, which is used for make format in c, you'll have to install that manually if you want to auto-format the C code)

Known good versions: GCC 12.1.0; Clang 14.0.0; Rustc and Cargo 1.67.1; .NET 7.0.200; GNAT 12.1.0; on Ubuntu 22.04. These correspond to the following pre-built Docker file: https://doi.org/10.5281/zenodo.7649567

Experiments

The benchmarking scripts for network functions, which are independent of the rest, are in benchmarking/.

The experiments presented in the paper, including replication instructions, are in experiments/. We've also provided the actual data collected on our hardware in experiments/results_example; you can rename the folder to results and run the scripts to plot it as per the instructions.