关于
I am a researcher at Azure Research – Security and Privacy in the Azure Office of the CTO. My research interests are in the area of computer systems with aspects to computer architecture and security. My current research centers around confidential computing; in particular, new forms of trusted execution and side-channel resilience. My pioneer research on confidential GPUs and confidential AI accelerators has defined a blueprint for accelerated confidential computing hardware, which has been adopted by the industry, including NVIDIA.
Prior to joining Microsoft, I completed my Ph.D. in Computer Science at Ecole Polytechnique Federale de Lausanne (EPFL) under the supervision of Babak Falsafi. There, I co-architected CloudSuite, a widely adopted benchmark suite for scale-out cloud services, and worked on processor architectures and memory systems for scale-out datacenters. CloudSuite’s micro-architectural characterization on scale-out servers laid the foundation for the first generation of Cavium ThunderX server CPUs.
Selected Publications
- Confidential Container Groups, In Communications of the ACM, October 2024.[Azure GA]
- Confidential Computing within an AI Accelerator, In USENIX Annual Technical Conference, July 2023.
- Graviton: Trusted Execution Environments on GPUs, In USENIX Symposium on Operating Systems Design and Implementation, October 2018. [Demo]
- Scale-Out Processors, In ACM/IEEE International Symposium on Computer Architecture, June 2012. Inclusion in ISCA 1996-2020 Retrospective.
- Clearing the Clouds: A Study of Emerging Scale-Out Workloads on Modern Hardware, In ACM International Conference on Architectural Support for Programming Languages and Operating Systems, March 2012. ASPLOS Best Paper Award. IEEE Micro Top Picks 2014. ACM ASPLOS 2023 Influential Paper Award.