Alessandro Morari, Ph.D.

About

I build AI systems across the full stack, from GPU kernel optimization to AI product leadership.

At NVIDIA, I lead efforts in AI-driven GPU kernel development and evaluate next-generation programming models for accelerated computing. Before NVIDIA, I led a team at IBM Research that shipped Watson Code Assistant from research through production, one of the earliest large-scale generative AI products. Earlier: Summit and Sierra supercomputers, distributed runtime systems at PNNL, and I created NYU Courant's first course on High Performance Machine Learning.

30+ publications. 15 patents. Ph.D. in Computer Architecture.

Education:

• Ph.D. in Computer Architecture, Polytechnic University of Catalunya.

• M.Sc. in Computer Engineering, University of Rome Tor Vergata.

• B.Sc. in Computer Engineering, Roma Tre University.

Awards

• Outstanding Division Achievement Award, IBM (2018) — CORAL Summit and Sierra Supercomputers

• Outstanding Performance Award, PNNL (2014) — Graph Engine for Multithreaded Systems (GEMS)

• IPDPS Best Paper Award (2012)

• IBM Top Technical Talent Award, 2017

• Exemplar People Manager Award, IBM

In the Press

CUDA Tile Programming Model (2025)

• "NVIDIA Introduces CUDA 13.1 with CUDA Tile". Calling it the largest advancement since the NVIDIA CUDA platform was inroduced in 2006, NVIDIA has launched CUDA 13.1 with CUDA Tile https://insidehpc.com/2025/12/nvidia-introduces-cuda-13-1-with-cuda-tile/

Summit/Sierra Supercomputers (2018)

• "Two DOE Supercomputers Top List of World's Fastest" - U.S. Department of Energy Oak Ridge's Summit and Lawrence Livermore's Sierra ranked #1 and #2 globally https://www.energy.gov/articles/two-doe-supercomputers-top-list-worlds-fastest

• "Sierra honored with Top Supercomputing Achievement" - HPCwire Editors' Choice Award Lawrence Livermore National Laboratory https://www.llnl.gov/article/44896/sierra-honored-top-supercomputing-achievement-hpcwire

• "Summit Supercomputer is Already Making its Mark on Science" - HPCwire Coverage of Summit's scientific impact and architecture https://www.hpcwire.com/2018/09/20/summit-supercomputer-is-already-making-its-mark-on-science/

GEMS massively multithreaded graph runtime (2014-2019)

• "Startup Trovares Brings HPC to Graph Analytics" - HPCwire Seattle startup commercializes PNNL graph database technology https://www.hpcwire.com/2019/04/16/startup-trovares-brings-hpc-to-graph-analytics/

• "A Meaningful Data Miner" - PNNL Science Highlights GEMS framework enables graph queries on billion-triple datasets https://www.pnnl.gov/science/highlights/highlight.asp?id=3949

• "PNNL: A Collection of Quality HPC Research at IEEE Cluster 2015" - PNNL High-performance dictionary encoding for RDF datasets https://www.pnnl.gov/science/highlights/highlight.asp?id=4018

Selected Papers

1. Scaling Semantic Graph Databases in Size and Performance

IEEE Micro, 2014

Alessandro Morari, Vito Giovanni Castellana, Oreste Villa, Antonino Tumeo, Jesse Weaver, David Haglin, Sutanay Choudhury, John Feo

A software stack that relies primarily on graph-based methods to implement scalable RDF databases on commodity clusters.

• DOI: https://doi.org/10.1109/MM.2014.55

2. In-Memory Graph Databases for Web-Scale Data

IEEE Computer, 2015

Vito Giovanni Castellana, Alessandro Morari, Jesse Weaver, Antonino Tumeo, David Haglin, Oreste Villa, John Feo

High-performance in-memory graph database system for web-scale semantic data.

• DOI: https://doi.org/10.1109/MC.2015.74

3. A Quantitative Analysis of OS Noise

IPDPS 2011

Alessandro Morari, Roberto Gioiosa, Robert W. Wisniewski, Francisco J. Cazorla, Mateo Valero

Developed technique to provide quantitative analysis of OS events that limit application scalability on large-scale supercomputers.

• DOI: https://doi.org/10.1109/IPDPS.2011.84

4. Evaluating the Impact of TLB Misses on Future HPC Systems

IPDPS 2012

Alessandro Morari, Roberto Gioiosa, Robert W. Wisniewski, Bryan S. Rosenburg, Todd Inglett, Mateo Valero

Quantitative analysis of TLB misses on Blue Gene/P at scale (up to 4096 cores).

• DOI: https://doi.org/10.1109/IPDPS.2012.94

5. Scaling Irregular Applications through Data Aggregation and Software Multithreading

IPDPS 2014

Alessandro Morari, Antonino Tumeo, Daniel G. Chavarría-Miranda, Oreste Villa, Mateo Valero

Techniques for scaling irregular applications on distributed memory systems.

• DOI: https://doi.org/10.1109/IPDPS.2014.117

6. D2A: A Dataset Built for AI-Based Vulnerability Detection Methods Using Differential Analysis

ICSE-SEIP 2021

Yunhui Zheng, Saurabh Pujar, Burn Lewis, Luca Buratti, Edward Epstein, Bo Yang, Jim Laredo, Alessandro Morari, Zhong Su

Production ML dataset for code vulnerability detection - shows ML deployment capability.

• DOI: https://doi.org/10.1109/ICSE-SEIP52600.2021.00020

Presentations and Seminars

NeurIPS https://nips.cc/Expo/Conferences/2020/talkpanel/21302

New York University https://cs.nyu.edu/dynamic/news/colloquium/1010/

HPCXXL https://hpcxxl.org/wp-content/uploads/2017/09/HPCXXL-Fall-2017-V6.pdf

PiPhilly https://www.piphilly.org/it/piphilly-events/video-galleries/ai-event-videos/

Links

Linkedin Profile

DBLP Profile

Arxiv Profile

Google Scholar Profile

ResearchGate Profile

Papers With Code Profile