Abdul Rafae Noor
PhD Candidate
Department of Computer Science
University of Illinois, Urbana Champaign
I am a 4th year Ph.D candidate at the Department of Computer Science at the University of Illinois, Urbana Champaign. My research interests includes using automated compiler techniques for code generation for heterogenous systems. I am advised by Prof. Vikram Adve.
Education
September 2016 - July 2020
Lahore University of Management Science
B.S Computer Science
Graduated on Dean's Honor List. Advised by Dr. Fareed Zaffar (Associate Professor, Dept. of Computer Science) and Dr. Junaid Haroon Siddique (Associate Professor, Dept. of Computer Science). Focus on compilers and program analysis.
August 2020 - Present
University of Illinois, Urbana Champaign
Graduate Research Assistant, Computer Science
Advised by Professor Vikram Adve, conducting research on heterogenous computing and retargetable tensor code generation frameworks.
Publications
2024
A. Kothari*, A. Noor*, M. Xu, H. Uddin, D. Baronia, S. Baziotis, V. Adve, C. Mendis, S. Sengupta "Hydride: A Retargetable and Extensible Synthesis-based Compiler for Modern Hardware Architectures." (* Equal contribution) To appear in (ASPLOS' 24)
2022
A. Ejjeh, A. Councilman, A. Kothari, M. Kotsifakou, A. Noor, H. Sharif, Y. Zhao , S. Adve, S. Misailovic, V. Adve "HPVM: Hardware-Agnostic Programming for Heterogenous Parallel Systems" in IEEE Micro [PDF]
2021
A. Ahmad, A. Noor, H. Sharif, U.Hameed, S.Asif, M.Anwar, A.Gehani, J.Siddique, F.Zaffar "TRIMMER: An Automated System for Configuration-based Software Debloating" in IEEE Transactions on Software Engineering [PDF]
News
November , 2023
Presented our work on HDC++ and HPVM-HDC at the University of San Diego, California. HDC++ is a high-level parallel programming model for expressing hyper-dimensional computing application which can be compiled through HPVM-HDC, a compiler infrastructure for heterogenous computing to enable targeting CPU, GPU, FPGA and hyper-dimensional computing accelerators.
April 25, 2022
HPVM is a retargetable compiler infrastructure for heterogeneous parallel systems that targets CPUs, GPUs, and FPGAs, and has been used for other domain-specific accelerators. HPVM uses a target-independent compiler IR that extends the LLVM compiler IR with an explicit, hierarchical dataflow graph representation that captures task, data, and (streaming) pipelined parallelism.
PhD Qualifying Exam
April 08, 2022
Successfully completed the Computer Science PhD Qualifying Exam at UIUC. The primary area was Compilers, with secondary areas of Parallel Computing and Computer Architecture.
2022
Each year, the Department of Computer Science recognizes the achievements, leadership, and potential of graduate students who have earned prestigious fellowships and awards. I received the prestigious Sohaib and Sara Abbasi Computer Science Fellowship.
