AUSTIN, Texas — Research into using artificial intelligence to improve the performance and energy efficiency of computer operating systems will be led by a team at The University of Texas at Austin, thanks to a major grant from the National Science Foundation's Computing Exploration Program.
Today's operating systems pose a significant barrier to many promising innovations in computer hardware and applications, from personal assistant robots to autonomous vehicles, to edge computing that enables smart cities and energy-efficient cloud computing. These operating systems often follow “one-size-fits-all” rules for how to allocate hardware resources among different applications running simultaneously. These rules are inflexible, making it difficult to integrate new advancements and leading to poor performance and inefficient use. A UT research team plans to use AI to tackle this problem.
“Learning-Directed Operating System (LDOS) employs AI-assisted intelligent resource management to automatically adapt as new applications and hardware emerge,” said Aditya Akela, first chair of the Computer Science department, who is leading the project. “This allows computing devices to be used at near-optimal efficiency while meeting the needs of any application, making computing infrastructure 'self-driving' by automating the implementation and management of the OS.”
Akela said the project goes beyond academia, bringing together computer scientists from the University of Texas, the Texas Advanced Computing Center, the University of Illinois at Urbana-Champaign, the University of Pennsylvania and the University of Wisconsin-Madison, as well as industry partners including Amazon, Bosch, Cisco, Google, Microsoft and Broadcom.
“These partners collectively develop and run the operating systems on much of the world's computing infrastructure, and we will work with them to create the next generation of open source, intelligent and adaptable OS,” Akella said. “We believe this project provides a unique opportunity to fundamentally change the trajectory of computing.”
He added that this new style of operating system could help autonomous robots become “the smartphones of the 2030s and beyond.”
“The smartphone revolution was made possible by new OS frameworks (such as iOS and Android) that allowed users to run any third-party app, which in turn determined how we interact with technology and essential services,” he said. “LDOS can similarly usher in an era of affordable personal robots and apps to assist us in our daily activities and improve society, especially for the elderly and disadvantaged.”
In addition to research initiatives, the project will create new undergraduate and graduate curricula that will include modules, courses, and certificates that explore the interplay between computer systems and AI. The project's initiatives to broaden participation aim to develop leadership among underrepresented groups in AI and prepare them for careers in AI and computer systems technology and research. These initiatives aim to benefit hundreds to thousands of students each year.
The project creates natural synergies with the University of Texas’ Machine Learning Institute, which is researching fundamental questions about new generative AI and machine learning techniques, and the Generative AI Center, which provides high-end GPU resources for training and evaluating new generative AI models.
Co-principal investigators from UT's Computer Science Department are Joydeep Biswas, Swarat Chaudhuri, Shuchi Chawla, Işıl Dillig, Daehyeok Kim and Chris Rossbach. Co-principal investigators from UT's Chandra Family School of Electrical and Computer Engineering are Alex Dimakis and Sanjay Shakkottai.
The grant is for $12 million over five years, with $9.3 million of that going to the University of Texas. The Expeditions grant is part of the largest investment ever provided by NSF's Department of Computer and Information Science and Engineering. The official name of the project is “NSF Expeditions in Computing: Learning Directed Operating System (LDOS) – A Clean-Slate Paradigm for Operating Systems Design and Implementation.”
This new support furthers the University’s Year of AI, an initiative that demonstrates UT’s commitment to developing innovations and preparing leaders to navigate the ever-changing environment brought about by AI.
