imcg / ava

Automatic virtualization of (general) accelerators.

Automatic Virtualization of Accelerators

AvA is a research system for virtualizing general API-controlled accelerators automatically, developed in SCEA lab at University of Texas at Austin. AvA is prototyped on KVM and QEMU, compromising compatibility with automation for classic API remoting systems, and introducing hypervisor interposition for resource management and strong isolation.

This repository is the main codebase of AvA. We host customized Linux kernel, QEMU, LLVM, and sets of benchmarks in separate repositories.

We are refactoring AvA code for better extensibility and developer-friendliness. The refactoring breaks a few functionalities at this moment, but they are coming back.

Setup code tree

git clone [email protected]:utcs-scea/ava.git
cd ava
git submodule update --init --recursive

Documentations

• Build and Setup
• Guestlib Configurations
• Virtualize (Your Own) New APIs
• Introduction to LAPIS

Install

System requirements

AvA was fully tested on Ubuntu 18.04 (Linux 4.15) with GCC 7.5.0, Python 3.6.9, Boost 1.65.x, cmake 3.19.1 and Protobuf 3.0-3.9. The system also works in Ubuntu 16.04 with additional care of Python 3.6 and Clang-7 installation for CAvA scripts, but we do not maintain the support anymore. We plan to migrate AvA to Ubuntu 20.04.

Hardware and API

The following hardware and APIs are virtualized with AvA (excluding manually implemented Python forwarding):

API framework	Hardware
OpenCL 1.2	NVIDIA GTX 1080 / AMD RX 580
CUDA 10.0 (driver)	NVIDIA GTX 1080
CUDA 10.0 (runtime)	NVIDIA GTX 1080
TensorFlow 1.12 C	Intel Xeon E5-2643
TensorFlow 1.14 Python	NVIDIA GTX 1080
NCSDK v2	Intel Movidius NCS v1 & v2
GTI SDK 4.4.0.3	Gyrfalcon 2803 Plai Plug
QuickAssist 1.7	Intel QuickAssist
Custom FPGA on AmorphOS	AWS F1

For the status of support, please check out Build and Setup document.

Related repositories

• Kernel: https://github.com/utcs-scea/ava-kvm
• QEMU: https://github.com/utcs-scea/ava-qemu
• LLVM: https://github.com/utcs-scea/ava-llvm
• Benchmarks: https://github.com/utcs-scea/ava-benchmarks

Publications

Yu, Hangchen, Arthur M. Peters, Amogh Akshintala, and Christopher J. Rossbach. "AvA: Accelerated Virtualization of Accelerators." In Proceedings of the 25th International Conference on Architectural Support for Programming Languages and Operating Systems, pp. 807-825. ACM, 2020.

Yu, Hangchen, Arthur M. Peters, Amogh Akshintala, and Christopher J. Rossbach. "Automatic Virtualization of Accelerators." In Proceedings of the Workshop on Hot Topics in Operating Systems, pp. 58-65. ACM, 2019.

Developers and contributors

Name	Affiliation	Role	Contact
Hangchen Yu	Facebook & UT Austin	Main developer	[email protected]
Arthur M. Peters	Katana Graph & UT Austin	Main developer	[email protected]
Amogh Akshintala	Facebook & UNC
Zhiting Zhu	UT Austin
Tyler Hunt	Katana Graph & UT Austin
Christopher J. Rossbach	UT Austin & Katana Graph & VMware Research	Advisor	[email protected]