George Amvrosiadis
Contact Info

George Amvrosiadis
Assistant Research Professor
Dept. of Electrical and Computer Engineering

Collaborative Innovation Center 2311
4720 Forbes Ave, Pittsburgh, PA 15213
 Email: gamvrosi (at) cmu (dot) edu

Administrative Contact:
Dawn Sienko
Collaborative Innovation Center 4103
Email: dsienko (at) andrew (dot) cmu (dot) edu
About

I am an Assistant Research Professor in the department of Electrical & Computer Engineering at Carnegie Mellon University with a courtesy appointment in Computer Science. I am also a core member of the Parallel Data Lab. I spend part of my time at Amazon S3 as a Visiting Academic.

My current research focuses on distributed and cloud storage, new storage technologies, high performance computing, and storage for machine learning.

For a quick overview of my team's most recent work, check out these links: For a more in-depth look, check out the Publications section, or for a better grouping, the individual project websites: Note to prospective Ph.D. students: Feel free to reach out, but I generally have a hard time keeping up with email. If you want to make sure I won't miss your application then: (a) apply through the ECE department, and (b) include my name in the list of faculty you are interested in working with. Best of luck!

Note to current M.Sc. students: If you want to try your hand at research, reach out with your CV and transcript! You can take a look at the project websites listed above to get a taste of what we do, but keep in mind that there might be ongoing projects that are not listed yet.

Students
Past students:
Teaching
Selected Publications

Full list of publications is available here.
  • "File Systems Unfit as Distributed Storage Backends: Lessons from 10 Years of Ceph Evolution" , A. Aghayev, S. Weil, M. Kuchnik, M. Nelson, G. R. Ganger, G. Amvrosiadis. To Appear in the 27th Symposium on Operating Systems Principles (SOSP), 2019. (Acc.Rate: 13%)
  • "Compact Filters for Fast Online Data Partitioning" , Q. Zheng, C. D. Cranor, A. Jain, G. R. Ganger, G. A. Gibson, G. Amvrosiadis, B. W. Settlemyer, G. A. Grider. To Appear in the International Conference on Cluster Computing (CLUSTER), 2019.
  • "Data Storage Research Vision 2025" , G. Amvrosiadis, A. R. Butt, V. Tarasov, E. Zadok, M. Zhao, et al. Report on NSF Visioning Workshop held May 30 - June 1, 2018.
  • "The Atlas Cluster Trace Repository" , G. Amvrosiadis, M. Kuchnik, J. W. Park, C. Cranor, G. R. Ganger, E. Moore, N. DeBardeleben. In USENIX ;login:, Volume 43, No. 4, Winter 2018.
  • "Scaling Embedded In-Situ Indexing with DeltaFS", Q. Zheng, C. Cranor, G. Ganger, G. Amvrosiadis, G. Gibson, B. Settlemyer, G. Grider. In Proc. of the International Conference for High Performance Computing, Networking, Storage and Analysis (SC 2018), November 2018. (Acc.Rate: 19%)
  • "On the diversity of cluster workloads and its impact on research results", G. Amvrosiadis, J. W. Park, G. Ganger, G. Gibson, E. Baseman, N. DeBardeleben. In Proc. of the 2018 USENIX Annual Technical Conference (ATC), July 2018. (Acc.Rate: 20%)
Service
Recent Talks
  • Thoughts on the Marriage of Systems and ML, Keynote at 2nd Workshop on Machine Learning for Computing Systems, Supercomputing 2020. November 2020.
  • Getting Zoned Storage out of the friend zone, Samsung. August 2020.
  • In-situ Data Processing and Zoned Storage, University of Toronto. October 2019.
  • Data Storage Research Vision 2025, USENIX FAST 2019. February 2019.
  • How I Learned to Stop Worrying and Love the Exascale, Los Alamos National Laboratory. August 2018.
  • Computing with Exaflops and Other Scary Stories, Veritas. July 2018.
  • How I Learned to Stop Worrying and Love the Exascale, Microsoft Research Redmond. June 2018.
  • Paving the Way for Exascale Computing, Hewlett-Packard Enterprise. February 2018.
  • Unobtrusive Storage Maintenance, Columbia University. September 2017.
  • Data Loss: Worse than Zombie Apocalypse?, University of Toronto. April 2016.
  • Opportunistic Storage Maintenance, Veritas Research Labs. October 2015.