The SAFARI Research Group is led by Professor Onur Mutlu. We are part of the Department of Information Technology and Electrical Engineering (D-ITET) at ETH Zurich.
Our group conducts cutting-edge research and education in computer architecture, computing systems, hardware security, bioinformatics, and software/hardware co-design. The major goal is to design fundamentally better computing architectures. Our work spans the boundaries between applications, systems, languages, system software, compilers, and hardware, with architecture at the core. We tackle many fundamental issues in high performance, energy efficiency, hardware security, fault tolerance, predictable systems, dependable systems, hardware/software cooperation, and genome analysis. We are especially excited about novel, fundamentally-secure and fundamentally-efficient computation, communication and memory/storage paradigms. Recent impactful examples of the group’s work include leading research into memory/storage systems and new computing paradigms, e.g., the discovery of the DRAM RowHammer vulnerability and the development of a comprehensive Processing-in-Memory paradigm.
Read our group's contribution to the 2021 D-ITET evaluation report, highlighting our research impact, teaching and key group contributions: SAFARI 2021 D-ITET Evaluation
We are always looking for highly qualified and motivated students and researchers. If you are an ETH student and are interested in doing a bachelor or master's thesis with us, or a semester project, please see our Theses and Projects page for potential topics.
If you are about to finish your bachelor or master’s degree or are already doing research and are interested in working with us at any level (internship, PhD, postdoc, visiting scholar), please visit our Work with us page.
Congratulations to Lara Lazier, a bachelor student in our group, and her Team Kassiopeia, who took home the prize for “Most Creative Policy Response Alternative” during this year’s Cyber 9/12 […]
Our new paper “NAPEL: Near-Memory Computing Application Performance Prediction via Ensemble Learning”, the first machine-learning-based prediction framework for Near-Memory Computing to be presented at the Design Automation Conference in June […]
Onur discussed root causes and solutions to RowHammer at his talk last week on “RowHammer and Beyond” during the NYU ECE Seminar: [Slides (pptx)] Also check out his new paper with Jeremie Kim: “RowHammer: […]
Onur gave an invited talk last week at IBM Research in Yorktown Heights, NY on “Using Commodity Memory Devices to Support Fundamental Security Primitives”. His slides are now online: [slides (pptx)] […]
We’ve just released the benchmarks for our recent paper: “Analysis and Modeling of Collaborative Execution Strategies for Heterogeneous CPU-FPGA Architectures”, in collaboration with Prof. Hwu’s IMPACT Research Group: impact.crhc.illinois.edu Link […]
Onur was in Abu Dhabi this week at the SRC-Mubadala-Khalifa Forum on The Future of Artificial Intelligence Hardware Systems and gave a keynote talk on “Intelligent Architectures for Intelligent Machines” His […]
Onur Mutlu gave a keynote talk last week on “RowHammer and Beyond” at the 10th International Workshop on Constructive Side-Channel Analysis and Secure Design (COSADE) in Darmstadt. His slides […]
The Google Faculty Research Awards for 2018 were recently announced, and our group received an award in the Mobile category! The Google Faculty Research Awards Program supports world-class, permanent faculty […]
Onur Mutlu gave the Distinguished Lecture on “Processing Data Where It Makes Sense in Modern Computing Systems: Enabling In-Memory Computation” at George Washington University, Washington, DC, on Feb 15 2019. […]
Onur Mutlu gave the keynote talk on “Accelerating Genome Analysis: A Primer on an Ongoing Journey” at the AACBB workshop on Friday Feb 16, held jointly with HPCA 2019. His […]
Onur Mutlu was the distinguished speaker in the Endowed Lectureship Series at the University of Texas at Austin ECE Department where he gave three public talks on “Memory Systems and Memory-Centric Computing Systems: […]
Our new paper “D-RaNGe: Using Commodity DRAM Devices to Generate True Random Numbers with Low Latency and High Throughput,” presented by Jeremie Kim at HPCA 2019, is now online. Jeremie […]