People > Profile
Franz Franchetti, PhD
- Adjunct Professor of Biomedical Informatics
- Kavčić-Moura Professor of Electrical and Computer Engineering, Carnegie Mellon University
- Associate Dean for Research, College of Engineering, Carnegie Mellon University
- Director, Engineering Research Accelerator, Carnegie Mellon University
Department of Biomedical Informatics (DBMI) | University of Pittsburgh School of Medicine
Biography
Dr. Franz Franchetti is the Kavčić-Moura Professor of Electrical and Computer Engineering at Carnegie Mellon University, where he also serves as Associate Dean for Research in the College of Engineering and Director of the Engineering Research Accelerator. He holds an adjunct appointment in Biomedical Informatics at the University of Pittsburgh and was a Senior Visiting Scientist at RIKEN, Japan (2019–2024). Dr. Franchetti is internationally recognized for his leadership of the SPIRAL project — a system for automatic generation of high-performance software and hardware implementations for mathematical transforms — and for his co-founding of SpiralGen, Inc. He has received multiple Gordon Bell Prizes and over 20 best paper awards at major international venues, and holds 4 patents. His biomedical informatics work focuses on applying high-performance computing to clinical and translational research.
Research Interests
Dr. Franchetti’s research develops algorithmic, compiler, and hardware methods for automatically generating high-performance code and hardware implementations from mathematical specifications. The flagship project, SPIRAL, enables extreme performance portability for FFT, signal processing, sparse linear algebra, and scientific computing kernels across CPU, GPU, and FPGA platforms. Current work focuses on LLM-guided code generation for scientific computing, fully homomorphic encryption accelerators, processing-in-memory architectures, and applications to biomedical imaging and computational biology through his adjunct appointment at Pitt DBMI.
Specialties
- High-Performance Computing and Code Generation
- SPIRAL: Automatic Performance Optimization
- Fast Fourier Transforms and Linear Transforms
- AI and Machine Learning for Scientific Computing
- Biomedical and Clinical Informatics Applications of HPC
Websites & Links
Education & Training
- HTL Matura (Engineering High School Diploma, with distinction), EngineeringHTBLuVA Wiener Neustadt, Austria1994
- Dipl.-Ing. (M.Sc., with distinction), Technical MathematicsVienna University of Technology, Austria2000
- Dr. techn. (Ph.D., with distinction), Computational MathematicsVienna University of Technology, Austria2003
- Postdoctoral Research Associate (Erwin Schrödinger Fellowship)Carnegie Mellon University, Pittsburgh, PA2004–2005
Awards & Distinctions
- Gordon Bell Prize (Peak Performance Award)SC2006
- Gordon Bell Prize FinalistSC2005
- Dean's Early Career FellowshipCarnegie Institute of Technology, CMU2012
- HPC Challenge Class II Award (Most Productive System)HPC Challenge2010
- Multiple Best Paper Awards (HPEC, DSLWC, ISPA)HPEC / DSLWC / ISPA2006–2014
- ACM TODAES Best Paper AwardACM2014
- City of Wiener Neustadt Culture Prize (Science Category)City of Wiener Neustadt2008
- Erwin Schrödinger FellowshipAustrian Science Fund FWF2004–2005
Representative Publications
Franchetti F, Low TM, Popovici T, et al. Spiral: Extreme Performance Portability. Proceedings of the IEEE. 2018;106(11). Special Issue.
Milder PA, Franchetti F, Hoe JC, Püschel M. Computer Generation of Hardware for Linear Digital Signal Processing Transforms. ACM TODAES. 2012;17(2):Article 15. (ACM TODAES Best Paper Award 2014).
Akin B, Franchetti F, Hoe JC. HAMLeT Architecture for Parallel Data Reorganization in Memory. IEEE MICRO. 2016;36(1):14–23.
Rivera J, Franchetti F, Püschel M. Floating Point TVPI Abstract Domain. PACMPL. 8PLDI, 2024.
Sadi F, Pileggi L, Franchetti F. High Performance Merge Sort. US Patent 11,249,720. February 15, 2016.
Research Grants
| Grant / Funding Agency | Role | Grant Number | Years |
|---|---|---|---|
| MAGNET: MAthematics, ComputinG, and NETworking for Resource-Efficient Computational Science DoE | CMU PI | 2025–2028 | |
| DURBAN: Enhancing Performance Portability in HPC Software with AI DoE | CMU PI | 2024–2027 | |
| COSMIC: Co-designing Scalable ML-aided in-array Computations DARPA | CMU Co-PI | 2024–2026 | |
| FFTX: A Co-Design Project for Fast Fourier Transforms DoE | PI | 2018–2023 |