AI-driven research rethinks computer operating systems

Education
Webp bkjml0uzuehqukkfc5oxneca70cd
University of Texas Executive Vice President and Provost Sharon L. Wood | University of Texas at Austin

AUSTIN, Texas — Research aimed at using artificial intelligence to enhance the performance and energy efficiency of computer operating systems will be spearheaded by a team from The University of Texas at Austin. This initiative is supported by a significant grant from the U.S. National Science Foundation’s Expeditions in Computing program.

Current operating systems present challenges to innovations in computer hardware and applications, such as personal assistant robots, autonomous vehicles, and edge computing for smart cities. These systems often employ "one-size-fits-all" rules for allocating hardware resources among different applications running simultaneously. This rigidity hinders the integration of new advancements, leading to suboptimal performance and inefficient resource use. The UT research team plans to address this issue through AI.

“Our project will employ AI-aided intelligent resource management and auto-adapt as new applications and hardware emerge,” said Aditya Akella, the Regents Chair in Computer Sciences #1 who is leading the project. “This will enable computing devices to be used at near-optimal efficiency while meeting the needs of arbitrary applications, and it will make computing infrastructure ‘self-driving’ by automating OS implementation and management.”

The project extends beyond academia, involving computer scientists from UT, the Texas Advanced Computing Center, the University of Illinois at Urbana-Champaign, the University of Pennsylvania, and the University of Wisconsin-Madison. Industry partners include Amazon, Bosch, Cisco, Google, Microsoft, and Broadcom.

“These partners collectively develop and run operating systems for much of the world’s computing infrastructure,” Akella stated. “We believe that the project offers a timely opportunity to fundamentally change the trajectory of computing.”

Akella also noted that this new OS could help autonomous robots become “the smartphones of the 2030s and beyond.” He compared this potential transformation to how new OS frameworks like iOS and Android enabled users to run third-party apps that revolutionized interactions with technology.

In addition to research initiatives, the project aims to create new undergraduate and graduate curricula featuring modules, courses, and certificates exploring computer systems' interplay with AI. These educational efforts are designed to cultivate leadership among underrepresented groups in AI fields.

The project aligns with UT’s Machine Learning Laboratory's ongoing work on generative AI techniques and its Center for Generative AI's high-end GPU resources for training new models.

Co-principal investigators from UT’s Department of Computer Science include Joydeep Biswas, Swarat Chaudhuri, Shuchi Chawla, Işıl Dillig, Daehyeok Kim, and Chris Rossbach. Co-PIs from UT’s Chandra Family Department of Electrical and Computer Engineering are Alex Dimakis and Sanjay Shakkottai.

The grant totals $12 million over five years; $9.3 million will go directly to UT. NSF Expeditions awards represent some of its largest investments by its Directorate for Computer and Information Science and Engineering. The full name of this project is NSF Expeditions in Computing: Learning Directed Operating System (LDOS) – A Clean-Slate Paradigm for Operating Systems Design and Implementation.

This support advances UT's Year of AI initiative aimed at fostering innovation in artificial intelligence.