Joshua Bakita
Assistant Professor of Computer Science
Research Interests
Professor Bakita's research interests include GPU Scheduling, Real-Time Systems and Operating Systems. Professor Bakita’s research has had a significant impact on the real-time systems community. His GPU management libraries and tools have been downloaded over 15,000 times, with adoption by institutions such as Carnegie Mellon University, Tsinghua University, Virginia Tech, National University of Singapore, TU Munich, and Alibaba, among many others.
Prior to joining MBZUAI, Professor Bakita received his Ph.D. in Computer Science from the University of North Carolina at Chapel Hill, where his research focused on enabling efficient and time-predictable real-time systems on GPUs through advanced scheduling and partitioning techniques.
He has published in all three of the top real-time systems venues—RTSS, ECRTS, and RTAS—and received the Outstanding Paper Award at RTAS for his work on GPU partitioning. His approach emphasizes practical, systems-level solutions, informed by experience working on GPU performance optimization at Waymo, GM Research, and Microsoft.
- Ph.D. in Computer Science, University of North Carolina at Chapel Hill, USA.
- M.Sc. in Computer Science, University of North Carolina at Chapel Hill, USA.
- B.Sc. in Computer Science, University of North Carolina at Chapel Hill, USA.
- Outstanding Paper Award, Real-Time and Embedded Technology and Applications Symposium (RTAS), 2023;
- Teaching Assistant of the Year, University of North Carolina at Chapel Hill, 2019
His current work continues to focus on improving the predictability and efficiency of GPU-based systems, with new efforts extending into consumer, cloud, and formally verified embedded systems.
Selected publications:
- J. Bakita and J. H. Anderson, "The Advantage of the GPU as a Real-Time AI Accelerator," Real-Time Systems (Journal), 2025.
- J. Bakita and J.H Anderson, "Hardware Compute Partitioning on NVIDIA GPUs," Proceedings of the 29th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS), pp. 54-66, May 2023. Winner, outstanding paper award.
- J. Bakita, J. H. Anderson, "Enabling GPU Memory Oversubscription via Transparent Paging to an NVMe SSD," Proceedings of the 43rd Real-Time Systems Symposium (RTSS), pp. 370-382, Dec 2022.