About
I'm a researcher at Huawei's Von Neumann Research Center in Zurich. I build performance models for early system design, focused on CPUs and memory. Currently, I'm most interested in optimizations that cross the usual boundaries between layers of the computing stack, and in how AI can help find them.
Much of my previous work looked at where a system's time and energy go and how to get them back. I did my Ph.D. at UPC in Barcelona with Antonio González and José-María Arnau, designing hardware for computer vision. I also visited the Edge Computing Lab at Harvard. I still collaborate with the ARCO group on ray-tracing hardware and co-supervise a Ph.D. student there.
I also review papers for the computer architecture community, including as a program committee member for MICRO 2026 and HPCA 2027.
Publications
-
MICRO2026
Aneto: Predicting System Performance by Exploiting Cross-Workload Regularity
-
MICRO2026
MVP: A Motion-Predictive Speculative Vision Pipeline with Non-Blocking Drift Correction
-
HPCA2025
IRIS: Unleashing ISP-Software Cooperation to Optimize the Machine Vision Pipeline
-
MICRO2023
δLTA: Decoupling Camera Sampling from Processing to Avoid Redundant Computations in the Vision Pipeline
News
- 2026Received a Huawei Director's Award for Arm SME kernel work in Huawei's production BLAS library.
- Two papers accepted at MICRO 2026.
- Serving on the program committee of HPCA 2027.
- Serving on the program committee of MICRO 2026.
Contact
Say hello.
raul.taranco@gmail.com